ORNL-4528.txt

 

   

Dw - /50 |

. -UC-80 — Reactor Technology

 

 

E '_: ', ‘_ TWO-FLUID MOLTEN-SALT BREEDER REACTOR
' DESIGN STUDY (STATUS AS 0|= JANUARY 1, 1958)"]‘ _‘

_-R Robertson 0 L Smith e
- .R.B Brlggs ~E s Bettis SRR

AL T e o

 

 

 

 

.'.'?,‘.-:_EOAK RIDGE NA'I'IONAI. LABORA'I'ORY
: operaled by

umou CARB!DE CORPORATION
L for the o ST
u s A'I'OMlC ENERGY commss:on -

. DISTRIBUTION OF THIS DOCUMENT IS UNLIMITED

 

 

 
 

 

&

 

 

Prlnted in the Umfed Sfu!es of Amerlca. Avalluble from Clearmghouse for Federol
Scientific and Techmcol lnformahon National Bureau of Stcndards, R

7 U s. Depurfment of Commerce, Sprlngfneid Virginia 22151 - s

Prlce° Prmted Copy $3 00; Mrcrof‘che $O 65 L S

 

 

- Thus report was prepared as on account of Government sponsored work. Neithe;‘ th'e_ Un'ifedVStefr:esr,_;

“mor. the Comm:sslon, nor-any person acting on behalf of the Commissuon." .

" As. used in the above, * ‘person acting on behalf of the Commission®’ andudes any employee oF .~
: contrucfor of the Commission, or empioyee ‘of such confructor, to the- extent fhof such. employee 1
or . conlracfor of the Commnss:on, or_employee of such confructor prepares, dlssemma?es, or>

: provvdes access to, any mformuhon pursucmf to hns employmenf or comruc! wnh Ihe Commussnon,

 

 

'_. LEGAL: NOTICE -

<AL Mukes ony wurrnnty or represenfuhon, expressed or lrnphod with respect to the uccurocy, ey

complefeness, of usefulness of the mformuhon contained in this report, or lhot the ‘use of

ony mformahon, uppamfus, meihod or process dlsclosed in thls reporf rnuy not mfnnge
pnvately owned rights; or ~ . s :

'B Assumes any habuhhes with respect to fhe use of, or for damuges resulhng from fhe use of-'—_' o

cny mformahon, apparatus, mefl\od or process disclosed in this report, .

or his emp!oyment wnth such contruc#or.

 

 

 

 

 

 

 
 

g ' ORNL-4528
' UC-80 — Reactor Technology

»

 

 

 

 

 

"
Contract No, W-7405-eng-26
|
REACTOR DIVISION
TWO-FLUID MOLTEN-SALT BREEDER REACTOR DESIGN STUDY (STATUS AS OF JANUARY 1, 1968}
R. C. Robertson 0. L. Smith
R. B. Briggs E. S. Bettis
A
-
v LEGAL NOTICE
This report was prepared as an account of work
sponsored by the United States Government. Neither
the United States nor the United States Atomic Energy
Commission, nor any of their employees, nor any of
their contractors, subcontractors, or their employees,
makes any warranty, express or implied, or assumes any
legal Hlability or responsibility for the accuracy, com-
pleteness or usefulness of any information, apparatus,
product or process disclosed, or represents that its use
would not infringe privately owned rights. :
AUGUST 1970 -
OAK RIDGE NATIONAL LABORATORY
QOak Ridge, Tennessee
operated by
UNION CARBIDE CORPORATION
~ for the
fi\ U.S. ATOMIC ENERGY COMMISSION
o
b=
w

DISTRIBUTION OF THIS DOCUMENT IS UNLIMITED
 

(A

»
 

o

 

a4/

CONTENTS
N 3T T U 1
1. Introduction ISR 2
2. Résume of Design Considerations . .........ueeuneeenrnneeeeernnreeeoennaresnnnnecesennenns 2
3, Materials ... i i e i e e et ettt et i 6
30 Gemeral ... s 6
B2 1 S 7
K 0 T 5 P13 - | ()20 . 11
34 Graphite ............oiiiiiiiiiia, e 12
3.5 Graphite-to-Metal Joints .. ... ...ttt ittt i i i it e 14
4. General Plant Description and FLOWSHEELS - -+« v e v eee e e e e e e e e e e e e e e e e et e et 16
0 T € 1T Y PP 16
B SIE ittt it i e e ettt i it e 16
43 Reactor Plant . ... ... i i i i it i it i i it e i et 18
44 TurbinePlant ................... e ettt etee e te et a et e e 26
4.5 Salt ProcessingPlant..................... e P 29
5. Major COmPOnents . ...........uinutiint it e 32
5.1 Reactor ....c.onininii 32
5.2 Fuel Salt Primary Heat Exchanger ............ e, S s 40
5.3 Blanket Salt Primary Heat Exchanger ................... T 42
54 SaltCirculatingPumps ...........c.ciieiennnnnan.. fereererenese e e eireeeaaaa. 44
55 OffGasSystem ................c.o.... Ceneeae . 48
56 DrainTanks ..........ccoiiiiiiinnnnnnn. feevsene e et aaeead e esaneaas 49
ST SHEAM GEMETALONS - . . e« v e s e ve e seene e ee e eaa e e e e e e e e e e e e e e e et e e e e 53
58 SteamReheaters....................... e e e hireaeas e et 53
- 59 Reheat Steam Preheaters S e et eeaenas 57
5.10 Maintenance . .......... ettt eeeenans e S s T 57
6. Reactor Physics and Fuel Cycle Analysis . ... eeenaen .............. 60
6.1 Optimization of Reactor Parameters .............. e 60
6.2 Useful Life of Moderator Graphite . . e e e et tiess et aaeaenaas 63
63 Flux Flattening ................. e, T 66
64 Fuel Cell Calculations . .. . .. ... e, SR e 67
6.5 Temperature Coefficients of Reactivity .......... ..., e 68
- 6.6 Dynamics Analysis 70
111
 

iv

7. Cost Estimates .. ............ £ttt eesenneanas et eieeetisnesenaneassnesannaseenaneaannaae 71

 

T GNEIAl « v v v e e e e, . 71
7.2 Construction Costs ...... v eeesatecensneea e et 71
7.3 Power Production Costs ........................ e it eeseseseseaenaes 73
Appendix.A': Cost Estimates ............. e beateacearaaeecaaceatarenetenonrrasenaan e ... 5

 

"

€A -
"

¥)

 

&/ »

 

TWO-FLUID MOLTEN-SALT BREEDER REACTOR DESIGN STUDY
(STATUS AS OF JANUARY 1, 1968) '

R. C. Roberts‘on . 0 L. Smith
'R. B. Briggs E. S. Bettis

ABSTRACT

A conceptual design study of e 1000-Mw(e) thermal breeder power station based on a two-fluid -
MSBR was commenced in 1966 as part of a program to determine whether a molten-salt reactor using

the thonum— U fuel cycle could produce electric power at sufficiently low cost to be of interest and

at the same time show good utilization of U.S. nuclear fuel resources. This report covers the progress

- made in the study up to August 1967, at which time the two-fluid MSBR work was set aside in order

to study a single-fluid MSBR concept. The latter became of interest at that time due to the discovery
that protactinium and other fission products could be separated from a uxanmm-and-thonum-bearing

_ fuel salt by reductive extraction into liquid bismuth.

The two-fluid MSBR is graphrte-moderated and: -reflected, with a 7L1F-BeF2-UF4 fuel salt

~ circulated through the core and a LiF—ThF4-BeF2 blanket salt circulated through separate flow
channels distributed throughout the core, as well as in a surroundmg undermoderated region. The

fissrons raise the temperature of the fuel salt to about 1300 °F and that of the blanket salt to about -
1250°F. Heat is removed from the salts in shell—and-tube heat exchangers to raise the temperature of a
circulating NaBF4-NaF coolant salt to about 1150 F The coolant salt transports the heat to steam
generators and ‘reheaters to prov:de 3500-psia 1000 FllOOO F steam for a conventronal turbine-
generator.

The conceptual desxgn was based on use of four reactors and the associated heat transfer systems
in a so-called modular arrangement to supply steam to a single turbine-generator, This made it

- practical to consider replaeement of an entire reactor vessel assembly after the core graphite received

its allowable exposure to neutrons, The total fluence at which it Was thought that addmonal graphite

- dimensional changes would become excessive was taken as 3 X 10% neutrons/cm (E > 50 kev), or -

about eight years of full-power operation, _
All portions of the systems in contact with the fluoride or fluoroborate salts would be fabnmted

" of Hastelloy N that has a small amount of titanium added to improve the resistance to radiation

damage. The graphite would be a specially coated grade having low gas permeability to xenon and
better resistance to radiation damage than conventional material. The two-fluid concept involves

' _ joining graphite core elements to Haételloy N tubing using a brazing process developed at ORNL.

The reactors and associated systems would be housed in concrete cells to provrde biological

~ shielding and double containment of all radxoact:ve materials.

Plant flowsheets and layouts were developed sufficiently during the study to give an indication of

- feasibility and to give a basis for cost estimates, but no optimization studies were made. Safety aspects

were considered throughout the design effort, but no formal safety analysis was completed. .
. Fuel and blanket salts would be continuously processed in a nearby cell to remove fission products
and to recover the bred product. The processing rate would correspond to removal of uranium and

~ protactinium from the blanket on a 3-day cycle and rare-earth fission products from the core on a

60-day cycle. Since no conceptual designs for the chemical plant were completed, cost estimates could

-not be on a definitive basis. The tentatively estimated fuel-cycle cost is about 0.5 mill/kwhr, which

includes the fixed charges and operating costs for the processing equipment, the fuel inventory charge,

- -and the credit for bred fuel Graphxte replacement costs, which are not included, would add about 0.2
) rmrlllkwhr

The tentatively estimated total construction cost of a lOOO—Mw(e) MSBR stat:on ‘based on the

- early 1968 value of the dollar, is about $141 per kilowatt. The power production cost for a privately

owned station, based on fixed charges of 13,7% and 80% plant factor, is about 4 mills/kwhr, The net

- thermal efficiency of the plant would be about 44 9%. ,

~ The off-gas, fuel processing, afterheat removal, “and mamtcnance systems needed further
investigation at the time the study was suspended, and the limited performance of the graphite

- undoubtedly restricts the design and imposes a maintenance penalty, but the study did not disclose

any aspects which indicated that major technological discoveries would be required to design a two-

¢ fluid molten-salt reactor power station. The major concern was whether mechanical failure of graphite
- tubes in the reactor core would cause the effective lifetime of the core to be slgmficantly less than the
eight years m\posed by the effects of irradiation on the graphite.

 
 

 

 

1. INTRODUCTION

The basic objective of the Molten-Salt Reactor Pro-
gram is to develop the technology for economical
nuclear power reactors that make use of fluid fuels
which are solutions of fissile and fertile materials in

 suitable carrier salts. A major goal is to achieve a

thermal breeder reactor based on the thorium-223U
fuel cycle that will produce power at low cost while
conserving and extending the nation’s fuel resources.
Conceptual design studies of a variety of molten-salt
breeder reactors for large plants are an important part
of this program. In August 1966 we published a survey
report, ORNL-3996,! in which we described briefly the
status of molten-salt reactor technology and the designs
of reactors and fuel processing facilities for 1000-Mw(e)

- power stations. This survey led us to conclude that the
- two-fluid reactor which separates the fuel and blanket
salts held the most promise for development as a -
breeder reactor. The modular version, consisting of four
reactor modules and associated intermediate systems -

supplying steam to one turbme-generator, was selected
for more detailed analysis.

The study of the modular design of a 1000-Mw(e)
plant was begun in the fall of 1966, and some of the
results were published in the MSRP progress reports,

' ORNL-4037,> ORNL4119,2 and ORNL4191.* Much

of the effort was spent on designs for the core and in
exploring the effects of radiation-induced damage to

graphite on the core designs. The plant layout, the cell.

designs, the drain tank systems, the nuclear character-
istics, the maintenance, and the cost estimates were also
examined in more detail than had been possible in the
earlier survey.

Considerable progress had been made in these studxes
when, in August 1967, encouraging information ob-
tained from research on the processing of moltensalt.
fuels indicated that protactinium and some fission
products could be separated from the uranium-and-

thorium-containing fuel salt of a one-fluid reactor by

reductive extraction into liquid bismuth. At about this
same time, nucleéar calculations indicated that a conver-
sion ratio greater than 1 could be achieved in a

 

1Paul R. Kasten, E, S, Bettis, and Roy C. Robertson, Design

Studies of 1000-Mw{e} Molten-SaIt Breeder Reactors, ORNL-
3996 (August 1966). _

2MSR Program Semiann,” Progr Rept. Aug 31, 1 966
ORNL-4037.

SMSR Program Semiann, Progr Rept. Feb, 28, 1967, ORNL-
4119,

*MSR Program Semiann. Progr. Rept. Aug 31 1967,
ORNL-4191,

one-fluid reactor of acceptable dimensions by increasing
the fuel-salt-to-graphite ratio in the outer regions of the
core. The one-fluid breeder is mechanically simpler than
the two-fluid breeder because it involves only one salt
stream, which contains both the fissile (*?3U) and the
fertile (thorium) constituents. Also, the one-fluid
breeder is a direct descendant of the one-fluid Molten-

_ Salt Reactor Experiment, which has operated well at
Oak Ridge National Laboratory. The attractive pos-

sibility of being able to progress in a direct path from
the MSRE to large thermal breeder reactors of similar

design led us to set aside the studies of two-fluid -
‘breeders to examine one-fluid breeder reactors in equal

detail. The studies of the one-fluid breeders were begun
in September 1967 and are continuing.

Although the one-fluid breeder has the desirable
features mentioned above, the fact remains that the
two-fluid MSBR is inherently capable of achieving a
significantly higher breeding performance. This feature
alone will sustain interest in the two-fluid system. It is
thus important to document the progress made in the
two-fluid breeder study before it was set aside. Present-
ing this information adequately is difficult, because
several months of studies of the one-fluid reactor have
changed some of our ideas about MSBR design, and
new data relevant to the two-fluid reactor have con-

~ tinued to come from the research and development

program. For example, the physical properties of the
salts have a profound influence on the design, yet many

of these properties are under continuous study and

adjustment. Some of the new information will be
mentioned briefly, but the reader should understand
that this report does not fully represent current ideas

‘and that some designs and conceptual drawings pre-

sented here would be considerably altered if they were

to be reexamined on the basis of today’s knowledge.
The studies upon which this report is based involved

personnel from almost all the divisions of ORNL, but

~ particularly those from the Reactor Division, Reactor
Chemistry Division, Chemical Technology Division, the

Metals and Ceramics Division, and the General Engi-
neering Division. A group composed of members of
these divisions, under the leadership of E. S. Bettis,
provided the conceptual designs and data which are
basic to the report.

2. RESUME OF DESIGN CONSIDERATIONS

Several basic considerations influenced our choice of
a two-fluid MSBR concept and many of the details of
the plant design. They are reviewed here to provide the

<
 

"

A\

reader with a better understanding of the de31gn that
evolved.

A simplified diagram of a two-fluid breeder reactor is
shown in Fig. 2.1. The core of the reactor consists of an
array of tubular graphite elements in the center of the
reactor vessel. A molten fuel salt is recirculated through

the graphite elements and through a shell-and-tube heat

exchanger by means of a centrifugal pump. A molten
blanket salt is similarly recirculated through the space
around and between the graphite pieces in the reactor
vessel and through an external heat transport circuit.
Heat generated in the reactor is transferred from the
fuel and blanket salts to a coolant salt in the heat
exchangers. The coolant -salt is recirculated through
steam generators where the energy is used to convert
the feedwater into superheated steam that drives a
conventional turbinegenerator to produce electricity.

The MSBR is a thermal breeder reactor that is
intended to attain the highest breeding performance
consistent with producing power at low cost. Our past
studies have indicated that a good measure of the

BLANKET-SALT FROM

PROCESSING PLANT . P
BLANKET-SALT l
ll . CRCULATION - . . _
BLANKET-SACT ; ¥ I .
TO PROCESS- = TUBES

ING PLANT

      
 
 
 
     
     

1125°F

COOLANT-SALT TO
STEAM GENERA-~

- REACTOR
TORS i

 
    
 
 
 
 

BLANKET-SALT
HEAT EXCHANGER

 

1eF N

/ N
COOLANT-SALT FROM MEO°F :
FUEL-SALT HEAT o t

EXCHANGER

 

 

performance of a breeder system is the total quantity of
fissionable material that must be mined in order to
provide the fissile inventory for a large nuclear power
system. This total ore requirement should be low. The
terms that describe the performance vary with the

~assumed growth rate of the nuclear electrical industry

and the types of reactors that precede and accompany
the breeders, but in the range of interest the per-
formance of a breeder is approximately proportional to
the product of the breeding gain G and the reciprocal
of the square of the specific inventory, 1/S?. The
“conservation coefficient” G/S? for MSBR’s can be
expected to be in the range of 0.02 to 0.10, where the
specific inventory has units of kilograms of fissionable
‘material per megawatt of electnclty and the breeding
gain is dimensionless.

A practical thermal breeder reactor can only be
fueled on the thorium-232U cycle, and it has a small
potential breeding gain. Typically, n for an MSBR is
'2.22 neutrons produced per neutron absorbed in fissile
material that is an equilibrium mixture of ??>*U and

ORNL-DWG 69-6868

g
2
8
8

GRAPHITE

= | _~~ HELIUM FROM

A

H~  OFF-GAS SYSTEM

FUEL~SALT CIRCULATION PUMP

FUEL-SALT FROM
-a— PROCESSING PLANT |

—= FUEL-SALT TO
PROCESSING
- PLANT

 

 

   
  
 
  
 

 

 

/
f
I E

— i

=
o J

1M1°F

  

AR

 

  

 

  
 
   
   
  

 

 

 

 

- %’*\W 'OOO‘,FJ | EOARKET- SS‘?QIE T
DRAIN LINE TO BLANKET- -— -
" SALT DRAIN TANKS HE}-_'%XSTO { HEAT EXCHANGER
' SYSTEM GAs SEPARATOR
© FUEL-SALT
HEAT EXCHANGER
- ' COOLANT-SALT FROM
DRAIN LINE 10 - STEAM GENERATORS.
- DRAIN TANKS * 850°%F - -

 

‘Fig. 2.1. Simplified Flow Diagram of Two-Fluid MSBR,
 

 

235, Absorptlon of one neutron in fissile material
and one in fertile material leaves 0.22 of a neutron for
losses to _moderator, carrier salt, leakage, Iughgr iso-
topes, protactinium, fission products, and structural ma-
terials and for absorption in thorium to produce the
gain in 233U,

Achieving high performance in a breeder depends on
keeping the parasitic absorption of neutrons and the
spec1fic inventory of fissile material low. Losses to
carrier salt, moderator, and structural materials and the
rate and cost of processing to keep the fission product
losses low all décrease with increasing concentration of
'uranium in the fuel salt and increasing inventory in the
reactor core. The specific inventory, however, includes
the iventory in the heat transfer equipment external to
the reactor vessel, in storage, and in the fuel processing
plants, so that the specific inventory and the total
inventory cost increase rapidly with increasing concen-
tration of uranium in the fuel salt. The breedmg gain
~and specific inventory must be balanced to obtain the
highest breeding performance (large G/S?) that is
consistent with producing power at low cost.
- The favored fuel salt contains about 0.2 mole % UF,,

of which about 70% is 223U and 35U, 23% is 234U,
and 7% is 236U. The uranium fluoride is dissolved in a
"LiF-BeF, (67-33 mole %) carrier salt. As shown in
Table 3.1, this salt has a liquidus temperature of about
840°F and good flow and heat transfer properties at the
working temperatures. It also has excellent thermal and

radiation stability and, with the use of ?Li, a low cross

section for the parasitic absorption of neutrons. A
ThF,-? LiF-BeF, salt (27-71-2 mole %), which melts at
about 1040°F, is a good choice for the blanket salt. The
physical properties of this salt are also shown in Table
3.1.
Although lithium and beryllium nuclei are good
moderators for neutrons, the moderating properties of
- the fluoride salts are not sufficiently good, when
compared with their neutron absorbing properties, to
build a thermal breeder without the use of other
moderator. Graphite is the best ‘material for this
purpose, because it has good moderation properties, a
low neutron absorption cross section, and good struc-
tral properties at high temperature and can be used in
direct contact with molten fluoride salts.
The design and performance of the reactor depend
considerably on the effects of fast neutrons on the

graphite. Neutron irradiation causes graphite to change

dimensions and its physical properties to deteriorate.
The life of the graphite is expected to be limited to
some total exposure to fast neutrons and therefore to
vary inversely with the maximum power density in the

core. Selection of a design power density for the core

must be based on a balance between the costs of fuel -

inventory, periodic replacement of the graphite, and
other factors that reflect on the net cost of the
electricity produced.

In order for the graphite to have an acceptable
radiation hfe we estimate that the maximum power
density should not exceed about 100 kw per liter of
core volume, With this limit on power density, the core
of a centralstation power reactor would have a volume
of several hundred cubic feet. This size is too large for

- the core to consist of graphite bars and highly enriched

fuel salt contained in a thin metal shell and surrounded

by a region of blanket salt. The critical concentration of

?23U in the fuel salt would be so low that the
absorptions in the carrier salt and the graphite would be

_excessive. Absorption of neutrons by the shell would
- further degrade the performance. '
The concentration of 233U in the fuel salt can be

naised to the desired level by dispersing blanket salt
throughout the core. This is accomplished by making
the graphite moderator in the form of tubular elements
and flowing the fuel salt through the elements and the
blanket salt around the elements. The core composition
is obtained by optimizing the relative volumes of fuel
salt, blanket salt, and graphite within bounds imposed
by limits on the concentration of thorium in the
blanket salt and by the engineering of the core.

" Results of many calculations have shown that the
combined neutron losses to fuel and blanket carrier

salts, the graphite moderator, and higher isotopes will
be near 0.11 in an optimized reactor, leaving 0.11 for
other losses and the breeding gain. Leakage losses are
reduced to a small amount by a thorium blanket of
reasonable thickness around the core. The losses due to
protactinium are kept small by keeping its concen-
tration in the blanket salt low. This is accomplished by
having a blanket of large volume at low neutron flux or
by removing the protactinium from the blanket salt on
a few-day cycle and allowing it to decay in the

processing plant. Xenon-135 must be removed from the

fuel salt on a few-second cycle, or the surfaces of the

graphite elements must be sealed to greatly reduce the
rate of diffusion of xenon into the pores. Most of the -

other fission products must be removed by processing

the fuel salt on a 30-to 50-day cyde. Limiting the total

of the above losses to 0.03 ‘to 0.07 appears to be
reasonable; this leaves a potential breeding gain of 0.04
to 0.08. |

A reactor with a breeding gain in this range and a
specific inventory of 1.5 kg/Mw(e) or less will have
good breeding performance. In order to have this low a
 

 

a)

”

r

 

specific inventory, the amount of 233U external to the
reactor core must be kept to a minimum. The heat
transfer circuit of the reactor must be closely coupled

‘to the reactor vessel, and it must have high perform-

ance. The fissile inventory in the blanket systems must
be kept small by extracting the bred 223U from the
blanket salt on a few-day cycle and making it available
for adding to the fuel salt to compensate for burnup.
Processing the fuel and blanket salts at the reactor site

_is necessary to avoid inventory in transport and storage,

and short cooling- time is important in reducing the
inventory in processing. The processes must be simple
and involve few changes in the physical or chemical
nature of the salts if they are to be carried out rapidly
and inexpensively. Fluorination to remove the uranium
as the volatile UF¢ followed by vacuum distillation to
separate the carrier salt from the rare-earth fission
products satisfies these requirements for processing the
fuel salt. Fluorination to remove the uranium or
extraction of protactinium and uranium into molten
bismuth can satisfy the requirements for the blanket.
With thorium blanket salt dispersed throughout the
core, the breeding gain is largely independent of the size
of the core, but this arrangement imposes several
conditions on the design. The first of these is that
graphite elements must be joined to metal-piping in the
reactor vessel. A perfect separation between the fuel
and blanket salts is not essential to the safety of the
operation, but the leakage must not be so great as to

~ put an excessive burden on the processing facilities.
Processing considerations lead to a preference for any
leakage to be blanket salt into fuel salt, and the leakage

must be kept below about 1 ft*/day in a 1000-Mw(e)
plant. Such a plant would have several hundred graph-
ite-to-metal joints. Our experience led us to choose
graphite-to-metal brazing as the method for obtaim‘ng
adequate leak-tightness.

The graphite elements for the core must be of a size
and shape that are within the capability of manufac-
turers to make and inspect for reasonable cost and with
good quality control. Isotropic material appears de-
sirable and may be essential from the standpoint of
irradiation effects. Thicknesses of sections must be
limited so that the temperature rise due to heating in
the graphite is not large. Effects of irradiation increase
with temperature, and stresses increase with tempera-
ture difference, so a large rise in internal temperature
could result in a large decrease in service life of the core
elements. Graphite tubés 6 in. or less in diameter and

~with a wall % in. or less in thickness appear to fulfill all -

these requirements.

Neutron irradiation produces substantial changes in
length of the graphite elements, and the difference in
expansion of the graphite and the metal parts of the
reactor vessel with temperature changes can also be

~ large. These effects must be accommodated without

overstressing the graphite. We propose to accomplish
this by making the graphite elements in the form of
concentric tubes connected to the reactor vessel at only
one end in order to provide freedom for axial expansion
and contraction. The fuel salt would flow in and out at
the same end of the elements, and the connections
would be to tube sheets at the bottom of the reactor
vessel to allow the salt to drain completely.

Because of the irradiation effects, the graphite tubes
will have to be replaced periodically. Also, one could
expect an occasional failure of a graphite element or a
graphite-to-metal joint from other causes. The reactor
vessel and internals will be highly radioactive after a
short time at high power, and with the graphite
elements brazed to a tube sheet in the bottom of the
reactor vessel, individual tubes could not be readily

- inspected or replaced. We concluded that the most

practical way to renew the graphite in the core would
be to replace the entire reactor vessel and its contents,
Suitable provisions would be required for remotely
operated tools and viewing equipment to cut, weld, and
inspect joints in the piping system. Provisions for

handling and disposing of spent reactor vessels would

have to be included in the plant.

The high melting temperatures of the salts make it .
necessary to preheat the reactor equipment to high
temperature before introducing the salts and to main-
tain the temperature when they are present. The special
problems of maintenance and inspection of the reactor
equipment after it has become radioactive led to our
proposals to install the reactor systems in heated cells,
which are comparable to large furnaces, rather than to
apply heaters and insulation to the vessels and piping.

In our studies of designs for molten-salt breeder
reactors, we are concerned primarily with power sta-

~ tions having outputs of 1000 Mw(e) or more. The

capacities of salt circulation pumps, heat exchangers,
steam generators, etc., needed for such plants are
greater than could reasonably be designed into single
units. In the 1000-Mw(e) MSBR design described in
ORNL-3996,! ‘we chose to connect four primary heat
removal circuits to one reactor vessel, to provide one
coolant and steam generator circuit for each primary
heat removal circuit, and to send the steam from all the
steam generators in the plant to one turbine.

 
 

 

Since the two-fluid breeder has a blanket of low 233U
and high thorium content around the core to capture
the leakage neutrons, reactors of this type can have
about the same breeding performance over a wide range
of size if the maximum power density in the core is
held constant. These facts, together with the special
problems and time required to replace a reactor vessel,
led us to consider a modular design for the two-fluid
MSBR in which. separate, but smaller, reactor vessels
would be coupled to primary heat removal circuits to
provide four autonomous reactor systems delivering
steam to -one -turbine-generator. This modular plant
would be slightly larger than the integral plant, since
four small reactor vessels with associated control sys-
tems would be substituted for the single larger vessel.
Otherwise the equipment in the plant would be the
same. The advantage would be that the plant could con-
tinue to operate at part load while one or two modules
were down for maintenance. We were sufficiently im-
pressed by this capability to make the modular concept
the basis for the: design studies described in later sec-
tions of this report. No analysis was made of the opti-
mum size for a module. We simply decided for the pur-
poses of this study to provide four modules in our
1000-Mw(e) plant.
~ All our designs for MSBR plants have fuel and blanket
circulation systems that are separated from the steam
system by an intermediate coolant system. If the steam
system were coupled directly to the fuel salt system by
means of a steam generator, any leaks in the tubes of
the steam generator would result in steam or water
leaking into the fuel salt. Reactions between water and
fuel salt would not be violent, but corrosive hydrogen
fluoride would be generated, and uranium oxide would
precipitate in the salt. Also, special provisions would
have to be included in the design to prevent the fuel
circulation system from being raised to the high
pressure of the steam system, Molten sodium, helium,
and other coolants have been considered for use in the
coolant system, but we prefer a molten salt. Sodium
reacts with the fuel salt to generate considerable heat,
precipitate uranium, and raise the melting point of the

salt. Helium does not react with the salt but must be

used at high pressure in order to obtain a good heat
transfer coefficient in the primary heat exchanger. At

best the heat transfer coefficient with gas is con-

siderably less than can be obtained with sodium or salt

“and results in an undesirably high inventory of fuel salt

and fissionable uranium in the reactor system. The
TLiF-BeF, coolant salt used in the MSRE is a good
coolant, but it costs about $1400 per cubic foot, and its
melting point is about 840°F. We would prefer to have

a less exi)ensive cobli_ng’ sélt with a lower melting point.
The salt NaBF,-NaF (92-8 mole %) costs only about

~ $60 per cubic foot, melts at 725°F, and is a favored '

candidate for use in the coolant system.,

Minimum operating temperatures for the MSBR are
set by the liquidus temperatures of the salts, and the
materials of construction are governed by the operating
temperatures and the properties of the salts, The
reactor fuel and blanket systems must be operated at
temperatures above about 1000°F, and the coolant
system must be operated above about 750°F. High
nickel alloys have good resistance to corrosion by
fluoride salts at high temperature and good creep
strength to about 1300°F. Since the temperature must
be high and the materials are expensive, we believe it

appropriate to couple the reactor plant {o a steam cycle

that is representative of the best current practice. The
3500-psia, 1000°F-throttle, 1000°F-reheat cycle that is
presently being specified for most new large fossile-
fueled plants was selected for use in our design studies
largely on this basis. The supercritical -cycle has the
added advantage that the feedwater to the steam
generators could be preheated to 700°F without much
loss in thermal efficiency by direct injection of super-
heated steam into the water. This procedure may be
necessary if use of feedwater at a more common
temperature creates problems in the steam generators
by freezing coolant salt on the tubes. (At subcritical

- pressures the Loeffler cycle employing a steam circu-

lator and mixing drum probably would have to be used
to attain the reqursrte hrgh-temperature entenng
stream.)

Finally, it is 1mportant to emphasrze that the desrgns
discussed here are based largely on current technology
and developments that we believe to be readily achiev-
able. The materials, processes, and performance factors
are developed sufficiently that no major inventions
appear to be reqmred to solve the technologrcal

-problems

3, MATERIALS
3.1 General

This ‘section briefly discusses some of the materials
which are unique to moltensalt breeder reactors. These
include the fuel, blanket, and coolant salts; the reactor
graphite; and the Hastelloy N used to contain the salts.
A brazed joint of graphite to Hastelloy N is also
described.

These, or similar, materials have been under study at
ORNL for many years, beginning with the ANP
1)

«)

 

program in the early 1950’s and continuing through the
MSRE program to the present. Specific evidence ‘has
accumulated that fluoride salt mixtures containing
fissile and fertile materials have the nuclear and physical
properties to make them suitable for use in a molten-
salt thermal breeder reactor. The salts possess suitable
liquidus temperatures and stability to temperature and
irradiation, The Hastelloy N, used to contain the salts,
and the graphite, which acts as the moderator, are
compatible with each other and with the salts. Except
for the graphite, which suffers irradiation damage, there
are no characteristics of the materials which sig-
nificantly limit the MSBR in the concept discussed
here.

The accumulated background of mformatlon on the

materials is too extensive to be covered fully in this .

report. References are made, however, to some key re-
ports that contain more complete information or bibli-
ographies.

3.2 Salts -

3.2.1 General

~Table 3.1 shows the salt compositions and physical
properties used in the two-fluid MSBR study. Recently
measured values of the physical properties are also

included where pertinent. (See Sect. 5.6.1 for estimates .

of volumes of salts in the systems.)

3.2.2 Fuel Salt

The fuel salt is a ternary mixture of ’LiF, BeF,, and
233yF, (68.5-31.10.2 mole %). A phase diagram for
the system is shown in Fig, 3.1, and the properties are
given in Table 3.1, :

The MSRE uses essentially the same fuel salt except
that it contains 5 mole % of ZrF,; to eliminate the
possibility of precipitating UQ, in the event of ac-
cidental contamination of the system with oxygen or
water. The zirconium addition is judged to be unneces-
sary for the MSBR in that the MSRE has been operated
for four years without contaminating the fuel salt,3™?
and the fréquent processing of the MSBR fuel should
keep the oxide content low,

The MSRE data also indicate excellent compatibility
of the salt with the Hastelloy N and graphite materials
in the system. The corrosion rate of the metal is less
than 0.2 mil/year, and the mechanical properties are
virtually unaffected by long exposure to the salt. The
graphite is not wetted by the salt mixture, and bulk
permeation by the salt is less than 0.2%, well below the
amount considered acceptable.

As indicated in Fig. 3.1, on cooling of the fuel salt in
the temperature interval from about 450°C (842°F) to
438°C (820°F), the compound 2LiF-BeF, precipitates
from the melt. At 438°C (820°F) the salt mixture

solidifies and produces a mixture of two crystalline

phases, 2LiF-BeF, (89 wt %) and LiF-UF,4 (11%). On
reheating, the mixture resumes its initial composition
and physical properties without change.

A considerable body of information exists to indicate
that the MSRE fuel salt is stable under irradiation and

 

SW. R. Grimes, Chemlcalr Research and Development for
Molten-Salt Breeder Reactors, ORNL-TM-1853 (June 6, 1967).

, 6 paul N. Haubenreich and J. R. Engel, “Experience with the

Molten-Sait Reactor Experiment,” Nucl. Appl. Technol., etc.
(see list of references).

7W. R. Grimes, “Molten-Salt Reactor Chemistry,” Nucl. Appl.
Technol. (see list of references).

Table 3.1, Physical Properties of Salts for Two-Fluid MSBR?

 

Blanket Salt

 

Fuel Salt Coolant Salt
Reference temperature, °F . 1150 1200 . 988
(unless otherwise noted) : _ _
Components TLiF-BeF,-UF, TLiF-ThF 4-BeF5 NaBF4-NaF
Composition, mole % |  68.5-31.3-0.2 71272 . 92.0-8.0

. Molecular weight, approx Lt 34 103 104
Liquidus temperature °F - 842 1040 . 700 (725)

- Density, p, lb/ft - o . 127 x6 27714 , o 125(121 at 850 F)
Viscosity, u, 1b ft ™! hr™? o 2713 38119 o _ 12 (4.6 at 850°F)
Thermal conductivity, k, ' . 1L5(0.8) . 1.5(0.6) C 13027

 Btuhe ! 7 °F 1 R R -

' Heat capacity, ¢, Btu1b™ °F” -1 | 0.55+0.14 0.22 +0.06 . 0.14(0.36) -
" Vapor pressure, torrs (mm-Hg) at 1150 °F - <01 <0.1 : ' 40 (252)

 

@The physical properties shown are those generally in use at the time the two-fluid reactor study was set aside. Values in paren-
theses are based on current information and are believed to be more representative.

 
 

 

1035 -

 
     

" ALL TEMPERATURES ARE IN °C

. E = EUTECTIC _ _
P PERITECTIC ) LiF *4UF,
[UF,] = PRIMARY PHASE FIELD o4

 

 

ORNL-DWG 66—7634

 

Fig. 3.1, Two-Fluid MSBR Fuel Salt — The System LiF-BeF-UF 4.

to temperatureswell above 800°C (1470°F). The MSBR
fuel should behave similarly. However, if irradiated salt
is allowed to freeze and cool below about 100°C

- (212°F), rédiqleis occurs with release of F,. This
reaction can be easily suppressed by maintaining the

salt above, say, 200°C (390°F). '
Fission products will be produced in a 2225-Mw(t)

MSBR at the rate of about 2.3 kg/day. The success of a

moltensalt reactor as a breeder hinges upon the ability

to process the fuel and blanket salts rapidly enough to
maintain the fission products at relatively low levels and
on keeping the costs of this processing low enough to

afford attractive fuel cycle costs. (The processing

aspects are more fully discussed in Sect.4.4.) Even with
rapid processing, however, the fission product concen-
trations are high enough to cause their behavior in the
salt to be of interest. S
 

L ¥

@

 

Table 3.2. Approximate Fission Product Distribution in MSRE After

 

 

32,000 Mwhr of Operation
Isotope l;v:;stoRrg Percent in Percent on Percent on Percent in
(dis/min)® Fuel G:aphiteb Hastelloy Nb Cover Gas?
CX 101 7
Mo 7.91 0.94 10.9 40.5 17
1324, 5.86 083 10.0 70.0 , 66
103pu 3.36 0.13 6.6 14.9 40
?5Nb 4.40 0.044 36.4 34.1 : 5.7
S7r  6.00 . 96.1 0.03 : 0.06 0.14
B9Sr 5.02 77.0 - 0.26 - 33¢
131 4.00 . 64.0 .09 169

 

9Total inventory calculated from the power history of the MSRE, : :
bValues for graphite and metal are based on the amounts found on specnmens removed from the core, and the values for the
cover gas are based on samples of gas obtained from the pump bowl

“Produced by decay of 8?Kr in cover gas.
dProduced by decay of 131Te.

Data obtained from the MSRE have confirmed the
chemists® predictions regarding the state of the fission
products in a moltensalt reactor. The rare gases
krypton and xenon are only slightly soluble in the

hightemperature salt and are readily removed by

sparging the salt with helium bubbles. Rubidium,

cesium, strontium, barium, zirconium, yttrium, and the
lanthanides form very stable fluorides, which are found
primarily in the salt. Some of these elements, such as

rubidium and cesium, have gaseous precursors and
appear in the graphite and the off-gas system in
proportion to the amounts of the precursors that escape

from the salt. The more noble metals from elements 41
and 42 (niobium and molybdenum) through element 52

(tellurium) are largely reduced to the metallic state in
the salt. They deposit on graphite and metal surfaces in
the reactor and somewhat surprisingly appear in the
cover gas, presumably as a “smoke” of metallic par-

" ticles. The distribution of representative fission

products of this group in the MSRE after 32,000 Mwhr
of operation is shown in Table 3.2. A similar distribu-
tion, modified to reflect differences in relative surface
areas and in flow conditions, must be expected in an

'MSBR. The data in Table 3.2 and other analyses of

samples of salt indicate that jodine forms stable iodides

" in the salt. Iodine found on MSRE surfaces and in the

cover gas is produced there by decay of the noble metal
tellurivm. Bromine forms stable bromldes that remain

_in the salt,

3.2.3 Blanket Salt _ _ |
_ The blanket salt for the two-fluid MSBR is a ternary

mixture of ?LiF, BeF,, and ThF, (71272 mole %).

| This system is Shown' in Fig. 3.2, and the properties are

listed in Table 3.1.
The blanket salt has a hqmdus temperature of about

- 560°C (1040°F), and during solidification the solid

phases LiF-ThF, and 3LiF-ThF, are formed, incorpora-
ting Be** in both the interstitial and substitutional
sites.’

‘The blanket salt can be expected to exhibit the same
good compatibility with Hastelloy N and graphite under
reactor conditions as does the fuel salt. Capsule tests in -
the MTR demonstrated the radiation stability of similar
salts containing ThF,. Early in the operation of the -

- MSRE there was some discussion of eventually opera-

ting with a fuel salt mixture containing thorium, but
this is now considered unnecessary since the results are
largely predictable.

The blanket salt will be processed on a rapid cycle to
remove the bred protactinium and/or fissile material in
order to minjmize the fissile inventory, the fission rate,
and the concentration of fission products in the blanket
salt.®. The chemical processing is discussed in more

“detail i m Sect. 4 4.

324 _Coolant Salt

In our 'design of an MSBR, a coolant is used for

~ transporting heat from the primary heat exchangers to

the steam generators ‘and reheaters. Characteristics
considered to be desirable in the coolant include low
melting temperature, compatibility with Hastelloy N,

 

8W. L. Carter and M. E. Whatley, Fuel and Blanket Processing
Development ' for Molten-Salt 'Breeder Reactors, ORNL-
TM-1852 (June 1967). ‘

 
 

 

  

10

’ ORNL-LR-DWG 37420AR6
ThE, 4444 \

 

 

W00
! TEMPERATURE IN °C
7 COMPOSITICN IN mote T
105
LiF-4The,
LiF-Thf, LiF-2Th A
3LIF-ThE, 55 4 iF-2Thiz 1000
LiF-2The;
P 897 950
2L|F-BeF2
LiF-ThE, 900
PTE2 850
P 597,
£ 568
3LIF-ThE
| a o
£ 565 %0
70
S 850
’%S’O OO 600 _
| <900 550 526
LiF - BeF
848 2LiF-BeF; 500450 400 400 450 500 543
| ' P asg £ 360 |

Fig. 3.2, Two-Fluid MSBR Blanket Salt — The System LiF-BeF;-ThF,.

resistance to decomposition by heat and radiation, good
heat transfer and pumping characteristics, low vapor
pressure at operating temperature, freedom from
violent chemical reactions with associated materials,
and low cost. Sodium is undesirable because of its
reactivity with air, water, and fuel salt. The MSRE
coolant, 7LiF-BeF, (66-34 mole %), has a liquidus
temperature near 455°C (850°F) and is more expensive
" than one would like to use in the large volume of an
MSBR system. Sodium fluoroborate of the eutectic
composition NaBF,-NaF (92-8 mole %) was selected as

most . nearly satisfying all the requirements for a

coolant. The phase diagram for the NaBF4-NaF system
is shown in Fig. 3.3, and the physical properties are
“given in Table 3.1.

Several mixtures of fluoroborates of the alkali metals »

were considered in making the selection. Some were
ruled out because of high viscosity or high cost.

ORNL-DWG 67-9423AR

 

LIQUIDUS
SOLIDUS

900 CRYSTAL
800

e

= 700

w

o

& 600

= : NaBF,

W (HIGH-TEMPERATURE. FORM)

2 500 +LIQUID -

|,—

"~ 400

300 + NoBF, (HIGH-

NaF 20 40 60 80 NaBF ,

NaBF, (mole %)

Fig. 3.3, TwoFluid MSBR Coolant Salt — The System
NaF-NaBF4. .
¥y

o

«)

)

 

11

Stoichiometric NaBF, does not exist in the molten

state without a very high partial pressure of BF; gas.

The eutectic composition, however, has most of the

* properties considered desirable for the MSBR coolant,

and it can operate with a dilute mixture of BF; in
helium at about 2 atm pressure as the cover gas. The
melting temperature of about 385°C (725°F) is ac-

ceptable. Although a lower temperature would be

desirable, it is not clear at this time whether the
liquidus temperature can be successfully depressed by
use of additives. The fluoroborate has a modest cost of
less than 50¢/Ib, and commercial grades may have
acceptable purity.

Thermal convection loop studies of the corrosion of

Hastelloy N by sodium fluoroborate at temperatures to
607°C (1130°F) have indicated a low corrosion rate in
the absence of contamination of the salt by moisture,

although not as low as with the MSRE coolant. As with
other fluoride salts, the presence of moisture greatly
increases the corrosion rates. The absence of severe
corrosion problems is confirmed qualitatively by ex-

Table 3.3. Nominal Chemica! Composition

 

 

of Hastelloy N
Standard Alloyl 1 Modified Alloy9
Elefnen ¢ (Much as Used  Recommended for
in MSRE) MSBR’s
(wt %)° (Wt %)
Nickel - Balance Balance
Molybdenum 15-18 12 .
- Chromium 6-8 7
Iron 5 0-4
Manganese -1 0.2-0.5
Silicon . 1 0.1 max
Boron 0.01 0.001 max
Titanium 0.5-1.0
Hafnium or Niobium ' 0-2
Copper , 0.35
Cobalt 0.2
Phosphorus ' - 0.015
Sulfur : 0.02 0.35
Carbon . 0,04-0.08
Tungsten : 0.5 :
Aluminum + Tltanlum , 0.5

perience with the circulation of sodium fluoroborate in

a large test loop for about 1800 hr. A corrosion product
precipitate, Na3CrF4, has been obtained from both

_ types of loops. Its solubility is inferred to be suf-

ficiently low that cold trapping may be required to
prevent the material from interfering with operation of
the coolant system by depositing on heat transfer
surfaces and in other cooled regions.

Molten sodium fluoroborate has been irradiated in
gamma fluxes as high as 8 X 107 r/hr without

~ significant effects on the salt or the Hastelloy N

container and specimens.”

3.3 Hastelloy N

The reactor vessel, piping, and primary and secondary
- heat transfer equipment require a material that is

resistant to corrosion by fluoride salts; compatible with
graphite; capable of being fabricated into complicated
shapes by rolling, forging, machining, and welding;
mechamcally strong and ductile at temperatures up to
700°C (1300°F); -and capable of maintaining these

 

%Single values are maxlmum percentages unless otherwise
specified.

the composition shown in Table 3.3 and was developed
in the ANP program to contain molten fluoride salts at
temperatures to about 870°C (1600°F). The MSRE was
constructed of standard Hastelloy N. The material was
obtained from commercial vendors, and it was fab-
ricated using conventional practices comparable with
those used for stainless steel. The major material
problem encountered was weld cracking, which was
overcome by slight changes in the melting practice and
by strict quality control of the materials. Heats of the
metal subject to cracking were identified and eliminated
by means of a weldablllty test mcluded as part of the
specifications.

Results of extensive corrosion tests, examination of
specimens exposed to the fuel salt in the center of the
core of the MSRE, and analyses of samples of fuel salt
from the MSRE have demonstrated the excellent

 resistance of Hastelloy N to corrosion by fluoride salts.
If the salt is kept slightly reducing and is not con-

properties during long exposure to this elevated tem-

perature in 2 neutron environment. Hastelloy N is the
preferred material for this application.

Hastelloy N is a mckel-base alloy contaxmng chro-
mium for oxidation resistance and molybdenum for

“strength at high temperature. The “standard alloy” has

 

9MSR Program Semiann. Progr. Rept. Feb. 29 1968, ORNL-

4254, p. 180.

tmua]ly contammated by oxygen or moisture, the
corrosion rate at temperatures to 700°C (1300°F) is
less than 0.5 mil/year. The effect of the corrosion is to

“gradually deplete the alloy of chromium, leaving behind

the major constituents and, at higher temperatures, a
network of subsurface voids.,

Irradiation of the standa_rd Hastelloy N by thermal
neutrons drastically reduces the ductility and stress
rupture life of the metal at high temperatures. This

 
 

12

Table 3.4. Physical Properties of Hastelloy N .

 

 

80°F 500°F 1000°F 1300°F 1500°F
Density, lblm 0.3209
Density, Ib/ft> ' 553.0 . - :
Thermal conductivity, But hr a1t 60 78 10.4 12.6 1441
Specific heat, Btulb ! °F ! : 0.098 0.104 0.1154 0.136 0.153
Coefficient of thermal expansion per °rb 57X10°%  70x10°¢ 8.6 X107° 9.5 %x107° 9.9 X 107°
Modulus of elasticity, Ibfin.2 31 X 108 - 29 X 10° c27x10%° - 25 x10° 24 X 10°
Electrical resistance, microhm-cm 120.5¢ 123.7 125.8 126.0%. 124,14
Approximate tensile strength, psi 115,000 106,000 95,000 75,000 55,000
Maximum allowable stress, psi¢ 25,000 20,000 17,000 3500

~ Maximum allowable stress, psi (bolts) . 10,000 7700 6600 3500

Melting temperature, °F 24702555

 

Taken directly from ref. 10. All other values found from interpolation of plots of ref. 10 data. See this reference for more pre-

cise information.

bAverage coefficient of expansion over 212 to 1832° F range is 8.6 X 10° per °F,

€Ref. 11.

deterioration results from the transmutation of 1°B to
lithium and helium, with the latter collecting in the

grain boundaries to promote intergranular cracking. The
irradiation effects become appreciable at a fluence of -
~about 10'® neutrons/cm?. At 650°C (1200°F) and

with stresses above 20,000 psi, metal irradiated to
fluences above 5 X 10'? can fracture with an elonga-
tion less than 0.5% and with less than 1% of the life of
unirradiated metal, Theoretical considerations and some
data indicate that the effects decrease with decreasing
stress.!® The damage occurs even though the boron
content of the alloy is as low as 1 ppm. It is, therefore,
not practical to limit the radiation effect by control of
trace amounts of boron,

Because the MSRE was intended to operate for only a
few years, the standard Hastelloy N was an acceptable
material of construction. A material with greater
resistance to radiation effects is, however, required for
those parts of an MSBR that are subjected to neutron
irradiation. Marked improvement in the properties of

irradiated Hastelloy N has been achieved by adding

small amounts of titanium and/or hafnium or niobium
to a shghtly altered base material to obtain modified
Hastelloy N of -the range of compos:tlons shown in

‘Table 3.3.

The stress rupture life and the ductility of modlfied
Hastelloy N can vary considerably with variations in
treatment and in amounts of some minor constituents.
In general, we have found that irradiation decreases the
rupture life and ductility of the modified alloy, but, for
irradiation to fluences of about 10?! neutrons/cm?

 

19y, E. McCoy, Jr., and 1. R. Weir, Jr., Materials Develop-

ment for Molten-Salt Breeder Reactors, ORNL-TM-1854 (June
1967)

(fast and thermal) at temperatures to 750°C (1380°F),
its properties are about equal to those of the standard
alloy when unirradiated. On this basis and on the
assumption that the reactor equipment would be made
of modified Hastelloy N, we used the extensive data on
the properties of unirradiated Hastelloy N in our studies
of designs for the reactor equipment. Those propertles
are reported in Table 3.4.

The specific heat, electrical resxstmty, and thermal
conductivity data all show inflections with respect to
temperature at 650°C (1200°F). This is thought to be

.due to an order-disorder reaction. No changes in the

mechanical properties are detectable as a result of this
reaction, however. The alloy has greater strength than
the austenitic stainless steels and is comparable with the
stronger alloys of the Hastelloy type. The maximum
allowable stresses shown in Table 34 were established
by performing mechanical property tests on experi-
mental heats of commercial size. The data were
reviewed by the American Society of Mechanical
Engineers Boiler and Pressure Vessel Code Committee,
and the stress values were approved for use under Case
1315 for Unfired Pressure Vessels and under Case 1345
for Nuclear Vessels.!?

73.4 "Graphite

The characteristics desired of the moderator material
for the core of a two-fluid MSBR concept are good
neutron moderation, low neutron absorption, com-

- patibility with the molten fuel and blanket salts and

 

“_American Society of Mechanical Engineers, Boiler and
Pressure Vessel Code, Section VIII, Unfired Pressure Vessels,
Case 1315, and Nuclear Vessel Construction, Case 1345,
 

n

[}

i)

1

 

low permeability to salt and gases, fabricability at

- reasonable cost, capability for being joined to Hastelloy -

N, and finally, ability to maintain all the desirable

properties after exposure to operating temperatures as

high as about 1400°F and to neutron fluences above

11023 neutrons/cm? (for E > 50 kev). In order to obtain

these characteristics a special grade of coated graphite
will have to be developed specifically for MSBR use.
The chemical purity and neutron performance, com-
patibility with materials, salt permeability, and strength
characteristics are sufficient in currently available

graphite. Preliminary experiments indicate that a sur--

face impregnation can be developed to keep the gas
absorption within acceptable limits. The effect of
neutron irradiation, however, is to first shrink and then

swell the graphite to cause an increase in porosity and,

we expect, a deterioration in physical properties. The
dimensional changes occur slowly, and their effects on
the neutronics of the reactor can be accommodated by
gradually adjusting the fuel-salt composition, although
at a small detriment to the nuclear performance. The
radiation damage to the graphite, however, limits the
useful life of the reactor core. Increases in cost that
result from more frequent replacements of the graphite
at higher power densities must be balanced against the

‘cost saving obtained from higher power density to

obtain a2 minimum cost for power.

The background of information on graphite is ex-
tensive. A detailed report ‘on graphite technology and
its influence on MSBR performance has been prepared
by Kasten et r:tl.12 A few factors are briefly reviewed
here.

Grade CGB!? was the first graphlte des1gned spe-
cifically for molten-salt reactor use and was first made
in commercial quantities for the MSRE. It is basically a
petroleurn needle coke bonded with coal-tar pitch,
extruded to rough-shape, and graphitized at 2800°C
(5072°F). High density and low gas permeability were
achieved through multiple pitch impregnations and heat
treatments. The material is highly anisotropic, however,
and while suitable for the MSRE. neutron fluence, it

~ would not have the d1mens10nal stabihty needed for an

MSBR.'*

 

12p R. Kasten et al., Graphite Behavior and Its Effects on
MSBR Performance, ORNL-TM-2136 (December 1968).

134 product of the Carbon Products Div1s1on of Umon
Carbide Corp. _
- 14g0e 1ef, 10 for other properties of MSRE graphite.

13

with Hastelloy N, sufficient strength and integrity to .
‘separate the fuel and blanket salts with good reliability,

Tests of the graphite indicate that isotropy is essential
if linear dimensional changes and overall volume

~changes are to be kept small in irradiated material. A

graphite with strong binder and a fairly high density
also appears to be important. For this reason, isotropic
graphite has been specified for use in the MSBR
concepts. Unless otherwise noted, this is the type of
graphite implied throughout this report. The nominal
physical properties expected of the graphite bef0re
irradiation are given in Table 3.5. ‘

There has been recent progress in the development of
isotropic and near-isotropic grades of graphite having
greater resistance to dimensional changes under irradia-
tion. Some of the sources of materials are Speer Carbon
Company = (grades 9948, 9949, 9950, 9972), Poco
Graphite, Inc. (grades "AXF, AXF-5Q, etc.), Carbon
Products Division of Union Carbide Corporation (grades

" ATJS and ATJ-SG), and Great Lakes Carbon Company

(grades H315A and H337). The isotropic graphites can
be made into various shapes by means of conventional
molding equipment, the limits on the gas permeability
playing a major role in the sequences of operations.
Much of the manufacturing information, however is
proprietary and unpublished.

While different grades of graphite behave somewhat
differently, it can be generally said that single graphite
crystals expand in the c-axis direction and contract in
the g-axis direction under irradiation by high-energy
neutrons. When large numbers of crystals are bonded
together to form a piece of commercial graphite, the
behavior under irradiation tends to be that shown in

Table 3.5. Nominal Values for

 

Properties of Graphite?
Density, lb/ft at room temperature ~115
Bending strength, psi 4000—-6000
~ Young's modulus of elasticity, £, psi - 1.7 X 106b

Poisson’s ratio, u : o 0.27¢
Thermal expansion, &, per F » 2.3 x107%¢.
Thermal conductivity, k, Btu hr ™~ ft'l OF"l 22-41¢
Electrical resistivity, ohm—cm X 10 : 8.9-9.9
Specific heat, Btu Ib OFT at600°F 0.33
Specific heat, Btu b F ! at 1200°F 042

 

2A specific grade of graphite and supplier had not been
selected for the two-fluid MSBR. Many of the graphite

' properties were, and still are, under investigation,

bp = =Eo + 1.2 X 10?T, where E = Young s modulus, psi, Eo =

- modulus at room temperature of 70 °F, T=°F.

“Poisson’s ratio is temperature independent.
Y= 383 X 107 +8.26 x1o‘9T—100x 10 “I“ where

r= °C between 400 and 1000°C.

ek =33 x 10 1;0 "i where T = °R between 550 and 4500° R
k=Btuhr™l ™! ) i
JRef. 10.

 
 

 

Fig. 6.4. Initially the volume contracts and the density
increases as some of the imperfections in the structure
are filled, On continued irradiation the volume increases
sharply, passing through the initial volume at a fluence
that decreases with increasing temperature. After ex-
amining the available data we concluded that a fluence

of about 2.5 X 10?2 neutrons/cm?® equivalent Pluto

dose could be sustained at 600°C (1112°F) without
deterioration of the physical properties of the graphite.
As explained in Sect. 6, this corresponds to a fluence of
5.1 X 10%? neutrons/cm? (E > 50 kev). For purposes

- of the design studies reported here, the time to

accumulate this dose was taken as the design hfetnme
for the graphite in the reactor core.

Graphite with a density of 115 lblft3 contains about
23 vol % voids. Low permeation of salt into the voids is
desirable to keep both the fission product poisoning
and the internal heat generation low, particularly after

. the reactor is drained, For the MSRE design we

specified that less than 0.5% of the bulk volume of the
graphite should fill with salt; specimens of grade CGB
graphite averaged less than 0.2%. The fuel and blanket
salts do not wet the graphite surface,'? and a pore size

- of less than about 1 u is sufficient to effectively keep

the salt out of the material. Experience with the MSRE
indicates that irradiation does not change this character-
istic.

Gaseous fissmn products tend to diffuse from the salt -

into the voids in the graphite. The graphite should have

a low gas permeability to reduce the levels of the xenon -

poison in the core and also to keep the heat generation
due to decay of fission product gases within the
graphite low. A target value of 0.5% xenon poison
fraction was selected for the two-fluid MSBR. The
permeability of graphite is usually measured with
helium at room temperature, and a value of less than
1077 cm?[sec is necessary if the diffusion of xenon at
reactor temperature is to be kept to an acceptable level.
Recent tests of six grades of isotropic graphite which

are of interest in the MSBR program showed per-

meabilities ranging from 3 X 107" to 1 X 1072
cm?/sec.’® Reducing the permeability sufficiently by

 

151, subsequent studies of one-fluid reactors. the design
lifetime was limited to a fluence of 3 X 10?2 neutrons/cm? (£
> 50 kev) on the basis that expansion of the graphite much
beyond the initial volume might increase the permeability to
salt and to account for the more rapld changes that occur at the
higher temperatures of 700 to 720°C in the graphite. More
recent data (July 1969) seem to confirm that the lower fluence
is a better value for graphite obtainable in the near future,

- Y6)MSR Program Semiann. Progr. Rept. Feb. 28, 1969,

0RNL-4396 (Aug. 1969).

14

pitch impregttation and gtaphitization treatments would

be very difficult; however, it is possible to achieve
acceptable permeabilities by depositing pyrolytlc
carbon in the surface pores.

In sealing the graphite with pyrolytlc carbon the
radiation-induced dimensional changes in the two ma-
terials may be sufficiently different to cause spafling of
the coating. This problem can be largely circumvented,
however, if the carbon is deposited in pores near the
surface rather than on the surface itself.” A method for
depositing the pyrocarbon has been developed at
ORNL. The graphite is cycled between a vacuum and a
regulated pressure of hydrocarbon (butadiene) gas while
it is being heated in a high-frequency induction field to
between 800 and 1000°C (1472—1832°F). The cycles
are of a few seconds duration, and permeabilities of less
than 1.3 X 1071° ¢m?/sec have been obtained.’” In -
one series of tests the depth of penetration at 800°C
(1472°F) seahng temperature was found to be about
0.015in..

Calculatlons were made by Kedl" to determine the

effect on the xenon poison fraction of sealingthe

graphite with a thin layer of pyrolytic carbon (or other
low-permeability graphite). Various xenon parameters

were chosen that would yield a high !3%Xe poison

fraction with ordinary graphite, and the calculations
were then extended to demonstrate the effect of the
coatings. The void fraction available to xenon was made
variable in such a way that it changed by one order of
magnitude when the diffusion coefficient changed by
two orders of magnitude. The diffusion coefficient of
1072 ft2/hr'® assumed for the bulk graphite is believed
to be readily attainable. The results are presented in

- Fig. 34. It is interesting to note that the diffusion

coefficient in the bulk graphite would have to be 1077
ft2 /hr or less in order to obtain a significant reduction

- in the xenon poison fraction, whereas an 8-mil coating

of 107 ft*/hr material would reduce the poison
fraction to the target value of 0.5%.

3.5 Graphite-toMeml Join_ts

- The two-fluid MSBR concept involves the joining of
the graphite core elements to stubs of Hastelloy N
tubing which are then welded into the tube sheets, as
indicated in Figs. 5.3 and 5.5. The graphite-to-metal

 

1IpmsR Program Semumn. Progr Rept Aug. 31 1968,
ORNL-4344 (Feb. 1969).

1871he dlffusxon coefficients given in an 3.4 are in £t2/hr for

xenon at 1200°F. These are numerically about equal to the ’

room temperature dlffusmn coeffic1ent for hehum ngen in
cm?/sec.
Ty

°

)

»

n

 

ORNL-DWG 68~ 553]

. SIGNIFICANT PARAMETERS: -
REACTOR POWER - 556 Mw{THERMAL)
CORE POWER DENSITY — 20 kw/liter
BULK GRAPHITE DIFFUSION COEFFICIENT -
1073 #2/hr _
BULK GRAPHITE AVAILABLE VOID—10%
CIRCULATING BUBBLE SURFACE AREA - 3000 f2

COMPOSED OF "ONCE THRU" BUBBLES AND
NO "RECIRCULATING" BUBBLES

MASS TRANSFER -COEFFICIENT TO BUBBLES—
. 20 ft/br -

FUEL CELL GEOMETRY —CONCENTRIC ANNULUS

 

XENON DIFFUSION COEFFICIENT IN COATING
(f12/hr)

 

 

'3 AVAILABLE VOID IN COATING (%)

 

 

 

 

 

 

 

 

 

 

3
S
=z
o
: N |
5 N : o 107 |
-'E;() A. "\\V \\ |
o e . i0-3 03
0 ’
0 5 10 15 20

COATING THICKNESS (mils)

Fig. 3.4. Effect of Pyrolytic Carbon Coating for Graphite on
Xenon Poison Fraction in Two-Fluid MSBR. '

joints would be made under carefully controlled shop
conditions. Methods for joining the graphite and Hastel-

loy are being studied at ORNL and have progressed

sufficiently to indicate that the matenals can be
successfully brazed together.

It is difficult to join graphite directly to Hastelloy
because the thermal coefficient of expansion of the
graphite is significantly lower than that of the metal.
The mean coefficient of thermal expansion of isotropic
graphite in the temperature range between 70 and
1100°F is about 24 X 107° in./°F, whereas that of

Hastelloy N is about 6.8 X 107 in./°F.° This dif-
ference is of primary concern when coohng from

brazing temperatures of about 2300°F.
One of the approaches to the problem is to design the
joint so that the Hastelloy N applies a compressive load

on the graphite as it cools, the graphite being stronger

in compression than in tension. Another approach is to
join the pgraphite to a transition material having a
coefficient of thermal expansion more nearly that of

the graphite. This material would in turn be brazed to
the Hastelloy N. A refinement of this is to use a series

of transition materials that would approach the thermal
expansion properties of the Hastelloy N in steps.

One of the families of materials investigated for use in.
transition pieces is the heavy-metal alloys of tungsten or
molybdenum, It was found that tungsten with nickel
and irori added in the ratio 7Ni/3Fe gave far better
fabrication characteristics than those with molyb-
denum.!” By adjusting the composition, the thermal
‘coefficient of expansion can be varied over the requisite
range of about 3 X 107 in./°F to 6 X 107 in./°F as
shown in Fig. 3.5.!7 Segments with highest tungsten
concentration would be located adjacent to the
graphite, and the segments with the most nickel and
iron would be next to the Hastelloy.? - -

Test specimens were prepared using nuclear-grade
graphite as well as the Poco type, which has a higher
coefficient of expansion, as shown in Fig. 3.5. The
distribution of the expansion coefficients of the in-
dividual segments in relation to those for the graphite
and the Hastelloy N is also shown in Fig. 3.5. The
composites were made by fabricating the segments
individually and copper-brazing them together in a
vacuum under light load. To achieve an effective bond
between the graphite. and ‘the metal requires prior
metalhzmg of the graphite surface by subjecting it to
gaseous products of a graphite-Cr,0; reduction re-
action conducted under a low vacuum at 1400°C."?

A minimum of intervening segments was employed in
an attempt to reduce fabrication costs. A typical
specimen consists of nuclear graphite, a 0.24in.thick
segment of Poco graphite, a 0.24n.-thick segment of

- 80% W-14% Ni—6% Fe alloy, and a 0.2-in.-thick

segment of 60% W—28% Ni—12% Fe alloy joined to the

ORNL—-DWG 68-14T66R

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T [T T T 1
100 | 5 ® DATA POINTS
- , e © PROPOSED COEFFICIENT OF
® | EXPANSION FOR THE SEGMENT |
2o % N JOINT
Zn
. we 80
8% 10 E W-Ni-Fe ALLOYS
Zz v E w
St L : =
EZ & T =
0
8= s0 S a > —
z o g N\ S
= < & N\ =
- w wi
. 50 -4 o =
o Q o
2. O <
< a I
40 '—lo o o—J
3 4 5 6 7 8 9 {0 @ 2 B

COEFFICIENT OF THERMAL EXPANSION
(m/m *c)x10~¢

Fig. 3.5. Expansion Coefficients of Transition Joint Materials
as a2 Function of Composition, Coefficients are mean values
between room temperature and 600°C and were determined on
an optical interferometer.

 
 

Hastelloy N.!® Extensive temperature cycling of the
specimens between 750°C and room temperature over a
20-min cycle failed to produce detectable cracks. The
copper bond between the metallized graphite and the
tungsten alloy remained intact, and there was no
evident reaction or alloying of the copper with the
chromium carbide. It was concluded that the graphite-
to-metal joint would give good performance under
MSBR service conditions.!®

4. GENERAL PLANT DESCRIPTION'
AND FLOWSHEETS

, 4,1' General

A large MSBR power station consists of three main
sections: the reactor plant that furnishes high-
temperature steam and breeds new fissionable material,
the turbine that generates the electric power, and the
chemical plant that processes the salts. The functions
and equipment for the three plants are closely inter-
dependent, but it is convenient to discuss them sepa-
rately. Less emphasis will be given to the description of
the turbine plant, since this involves more or less
standard equipment.

In the 1000-Mw(e) power station described in this
report, the heat is generated in four reactors, each
designed for a thermal output of about 575 Mw(t).
Each reactor. module is distinctly separate from the
others, having its own reactor vessel, primary fuel and

DISCHARGE

WASTE STORAGE
WASTE GAS
CHARCOAL ADSORBER
WASTE GAS TREATMENT
. AND STACK
TRANSFORMERS

PARKING

16

blanket heat exchangers, salt-circulating pumps, steam
generators, and steam reheaters. One chemical process-

. ing plant serves all four reactor modules. The steam

 
 
 

INTAKE

 
 

provided by the four modules supplies one 1000-Mw(e)
turbine-generator in the turbine plant. One regenerative

feedwater system consisting of two parallel streams.

returns boiler water to the reactor modules.

The reactor plant was the main subject of these design
studies, and very little was done on the site and building
layouts and on the chemical processing in addition to
what was reported in ORNL-3996.! In the interest of
making this report more complete, we have included
some information from ORNL-3996.

4.2 Site

The plant site is that described in the AEC handbook
for estimating costs. It is a 1200-acre plot of grass-
covered, level terrain adjacent to a river having adequate
flow for cooling-water requirements. The ground eleva-
tion is 20 ft above the high-water mark and is 40 ft
above the low-water level. A limestone foundation
exists about 8 ft below grade. The location is satis-
factory with respect to distance from population
centers, meteorological conditions, frequency and in-
tensity of earthquakes, and other environmental con-
ditions. - ' |

As shown in Fig. 4.1, the plant is in a 20-acre fenced
area above the high.water contour on the bank of the
river. The usual cooling-water intake and discharge

structures are provided, along with fuel-oil storage for a

ORNL-DWG 66-T133A

ELEVATION O fi
NORMAL RIVER LEVEL .

20 ft
ft

   
 
   
  
 
 
  
 
 
 
 

  

ft
80 ft

FUEL OIL STORAGE

 
 

STATION

O

   

RAILROAD
WATER PURIFICATION

 
   

WATER STORAGE
: (;ONDENSATE STORAGE

O
®
TURBINES

WELL

Fig. 4.1, Plant Site Layout.
 

g . startup boiler, a water purification plant, water storage

| tanks, and a deep well. This plant area also includes

" systems for treatment, storage, and disposal of radio-
active wastes. Space is provided for transformers and
switchyard. A railroad spur serves for transportation of
heavy equipment.

Ty

 

- - 530 ft

' °=

17

One large building houses the reactor, chemical
processing, and turbine plants, offices, shops, and all
supporting facilities. One version of this building,
shown in Figs. 4.2 and 4.3, is 250 ft wide and 530 ft
long; it rises 98 ft above and descends 48 ft below grade
level. An alternative plan in which the building is 340 ft

ORNL-DWG €6-7138A

 

)

ROLL-UP DOOR
R

 

HOT SHOP
AND STORAGE

ELECTRICAL EQUIPMENT AND
BUILDING VENTILATION SYSTEM

35 ft

 

 

 

 

 

 

 

 

 

 

 

 

 

r ———————————————————— = |
| | >
: i LP |
B ! I TURBINE -
I - '
{ REACTOR r - 13011
. i ' [
i i
; : HP AND IP 2501t
| | TURBINE
o
-
5 FEEDWAT
OFFICES, ' ER SYSTEM on
TREATMENT ROOM 1
LUNCH ROOMS,
. AND ETC.
3 FLOORS HIGH SHOPS 55 f
170 1t x 85 1 335 ft x 551 X 30 ft HIGH !

 

 

 

 

 

 

 

-

'cg-l'

. Fig. 4.2. Building Plan.

  

o= :

TURBINES .

*

—

‘ } Fig. 4.3. Building Elevation.

ORNL-DWG 66-T137A

TOP
OF RAIL

  

146 ft

      

f1 ABOVE NORMAL
RIVER LEVEL

 

 
 

 

18

ORNL-DWG 70-2tT4

+

290

 

TURBINE ROOM
125 H

 

FEEDWATER HEATERS
" 80H

4
4

 

 

REACTOR CONFINEMENT
STRUCTURE

BOH

 

TWO LEVELS
20H

CONTROL ROOM
SHOP—80 H
SHOP—-20H

 

 

 

 

e e e e

 

 

-—50---——————‘ 165 et 50 -ptet——— ({5 ——— )

OFFICE-TWO LEVELS
20H

 

 

 

4.3.1 Flowsheet

NOTE: DIMENSIONS IN FEET

Fig, 4.4. Alternative Building Layout (Used in Cost Estimate).

by 380 ft, as shown in Fig. 4.4, was used in estimating

the building costs. In either case, the building is of steel
frame construction with steel roof trusses, precast
concrete roof slabs, concrete floors with steel gratings
as required, and insulated aluminum or steel panel

A
8
4

walls. The wall joints are sealed in the reactor end of

the building to provide a confinement volume in the
event of a release of radioactivity. The reactor area is
provided with a separate ventilation and air-filtration
system that discharges to a stack.

4.3 Reactor Plant

A flowsheet for a reactor module is shoWn in Fig. 4.5,

In brief, the fuel salt enters the bottom of the reactor

vessel at a rate of 25 cfs at 1000°F, passes through the

core, and leaves at about 1300°F. It then enters the fuel

salt pump at the top of the primary heat exchanger,

where it is pumped into the center section of tubes.
After réversing direction at the bottom, the salt flows
upward through the outer section of tubes and into the
return line to the bottom of the reactor.

The fuel salt pump and its sump, or pump tank, are
below the reactor vessel, so that failure of the pump to
develop the required head causes the salt to drain from
the reactor vessel through the pump tank to the fuel
salt . drain tank. The tank above the pump impeller is
required during startup so that the fuel salt can be
pressurized from the drain tank into the primary system
to provide the pump with the necessary submergence
and surge volume as it starts and fills the reactor core.

Helium is used as the cover gas over the salt in the
pump bowl and as the medium for stripping gaseous
fission products from the salt. For this latter purpose,
small bubbles are injected into the salt in the suction
line to the pump and are removed with their burden of
krypton and xenon in a centrifugal separator in the line -
from the outlet of the heat exchanger to the reactor
vessel. This gas is circulated through a gaseous fission
product disposal system, described in Sect. 5.9.

The blanket salt enters the reactor vessel at a rate of
43 cfs at 1150°F. It flows along the vessel wall,
through the interstitial spaces between the graphite
elements of the core and the radial blanket, and exits at
about 1250°F. The fertile salt then flows into the
suction of the blanket pump and is pumped through the
blanket heat exchanger and back to the reactor vessel.
Helium covers the blanket salt at the salt-to-gas inter-
face in the pump. Only a small fraction of the fissions
occurs in the blanket, so there is no need for a gaseous
fission product removal system,

The sodium fluoroborate coolant salt is circulated to
the bottom of the fuel salt heat exchanger at a rate of
37.5 cfs at 850°F, flows upward through the shell, and
leaves at about 1111°F, It then flows through the shell
of the blanket salt heat exchanger, where it is heated to
about 1125°F, and returns to the coolant salt circu-
lating pump, where its pressure is raised from about 110
psig to 260 psig. The pump supplies about 87% of the
coolant salt to the steam generators and the remainder
to the steam reheaters. A cover-gas system is required
for the coolant circuit, the cover gas being 2 mixture of
boron trifluoride in helium. There is no requirement for
injecting cover gas into the circulating salt or for
removing it.

Each of the salt circulating systems is prowded with
heated drain tanks for safe storage of the salt during
shutdown of the reactor. These tanks are described in
detail in Sect. 5.6. The fuel drain tanks have cooling
systems for removal of afterheat. Flow of salt to the
tanks during a drain is by gravity ; salt is returned to the
systems from the tanks by pressurizing the tanks with
helium. A salt seal is frozen in the special valves in the
drain lines to effect a positive cutoff.
»

z

g _-

o

al

W

B

Zz

g:: 1234
= MODULE
e HEADERS

     

 

 

. ORNL-OWG 67-3084AR

   
  
     
  
 
   

BLANKET SALT
HEAT EXCHANGER
AND PUMP

FUEL SALT
HEAT EXCHANGER
COOLANT
AND PUMP : BLANKET
REACTOR : SALT .
~ PROCESSING

FUEL
PROCE SSING
4 f1¥

    
   
  

|

4

  
    

 

|,

 
   
  

 
  
     
   
 
   
 
  

  

PRODUCTS
DISPOSAL
SYSTEM

A

 
   

SYSTEMS

 
     
   

SUPERHEATER

  

REHEAT

 

 

PREHEATERS MIXING
‘ . . PUMP TEE
FUEL SALT " BLANKET SALT COCLANT SALT
DRAIN TANKS _ SALT  DRAIN TANK  DRAIN TANK
DRAIN TANK _ _
 FUEL _ - COOLANT STEAM PERFORMANCE DATA
POINT ft¥sec psig TEMP (°F) ~ POINT f1¥/sec psiq TEMP {°F) POINT 10°Ib/hr psia TEMP CF) 1000 Mwl(e) GENERATION PLANT
() 20 13 1300 @ 375 129 1125 2.52 3600 1000 NET OUTPUT 1000 Mwie)
(@ 250 9 1300 @) 375 110 1125 G 1007 3600 1000 ~ GROSS GENERATION  1034.9 Mw(e)
(3@ 250 146 1300 @ 375 260 1125 69 7.5 3515 1000 ~ BOILER FEED PUMPS - 29.4 Mw
(® =250 10 ’ 33 so0 208 1125 63 292 3515 1000 . BOOSTER PUMPS 9.2 Mwle)
(5 =250 50 1000 @ s0 212 850 292 3500 866 STATION AUXILIARY 25.7 Mwle)
(6) 250 31 1000 " (@ 81 252 25 639 513 €0 552 REACTOR HEAT INPUT 2225 . Mw(t)
(@ 250 18 1000 @ 81 194 850 €9 513 570 650 NET HEAT RATE 7601 Btu/kwhr
@) 35 203 850 69 128 570 650 NET EFFICIENCY 449 %
" BLANKET (3 375 198 850 128 540 1000 ' LEGEND ’
@ 43 125 1250 375 164 1114 513 540 1000 ' FUEL e
@ 43 80 125 @ 375 138 1N 500 172 706 BLANKE; —
, COOLAN ‘
@ 43 - 1110 1250 430 5inHg 92 . STEAM ——————
@) 43 200 %0 € 716 3500 559 WATER —— e —
@) 43 145 150 1007 3475 695 FREEZE VALVE B
- €9 1007 3800 700 DATA POINT
252 3800 700

Fig. 4.5, Flowsheet for 250—Mw(e) MSBR Modute.

 

61

 
 

 

 

A small side stream of fuel salt is taken from the fuel
systemn at the circulating pump discharge. After storage
in a transfer tank, the salt is processed and reconsti-
tuted in the associated chemical plant to remove fission
product contaminants and to adjust the composition.
The clean salt is retumed to the circulating system at
the pump bowl. A side stream is removed from the
blanket system, similarly processed for removal of bred
233p3 and 233U, and returned to the reactor. The
flowsheets for fuel and blanket salt chemical plants are
described in Sect. 4 4.

4.3.2 General Layout of Reactor Plant

The reactor plant consists of four major cell com-
plexes, as shown in Figs. 4.6 and 4.7, all contained in a

  
 

INSTRUMENTATION CELL

  

 

 

o=

I
I
{
[
i
|
I
I
I
I
|
|
1
I
|
|
I
L

 

 

OFF GAS PROCESS

CHEMICAL PROCESSING

STORAGE

reinforced concrete structure having outside dimensions
of about 150 by 170 ft and 45 to 65 ft high. Each
major cell complex includes a reactor cell, a coolant
cell, and a hotstorage cell to house a spent reactor
assembly. Two drain-tank cells and two off-gas cells are
located between the main cell complexes, and each
serves two reactors. A centrally located instrumentation
cell houses equipment for all four reactors. The

- chemical processing cell and the hot cells needed for

maintenance of radioactive equipment are also integral
parts of the structure.

All cells have removable top plugs of reinforced
concrete to permit maintenance operations to be
performed from above by use of remotely operated
tools and equipment. -

ORNL—DWG 68—28A

COOLANT SALT PUMP

BLANKET AND COOLANT DUMP TANKS

  

   

S

  
 
 

GENERATOR

    

  

O000000

DUMP TANKS
BLANKET HX

HX -

REACTOR

FLUSH SALT TANK

  

HOT

 
 
   
  
  

OFF GAS PROCESS ROOM -

O 5 10 15 20 25 30 35 40
HOT CELLS 1 ¢ 1 t t r 1 1|
' FEET

Fig. 4.6. Plan View of Reactor Plant.
 

»

(2]

¥}

21

CRNL—DWG 68 —-25A.

FUEL AND BLANKET PUMP COOLANT
DRIVE MOTORS SALT PUMP
STEAM GENERATOR \ .

 

 

 

 

 

 

 

 

REACTOR

 

 

HOT STORAGE

 

 

 

 

 

 

 

 

 

 

 

 

 

-----
BOSCERRY

 

     
 

  

 

 

 

 

 

 

 

2D
Dtenn

PiPING
s

 

 

 

 

 

REHEATER-|

 
 
   
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

BLANKET HEAT EXCHANGER AND PUMP

   

c. 5 10 15
bossdaabennd
FEET

 

 

TS PRIMARY HEAT EXCHANGER AND PUMP

 
 

THERMAL SHIELD

Fig. 4.7. Sectional Elevatibn of Reactor Cell,

All the cells containing fuel, blanket, and coolant salts
are provided with electric resistance heating elements
which preheat the systems and maintain the cell
ambient temperature at about 1100°F, well above the
liquidus temperatures of the salts. In addition to the
massive concrete biological shielding, the reactor cell
has thick double-walled steel liners to protect the
concrete from excessive temperatures and radiation-
induced damage. The liners also seal the cell spaces to
provide containment for all equipment which contains
radioactive material. The cell structure itself is housed
in a sealed confinement building which provides yet
another. line of defense agamst the escape of fissmn
products.

4.3.3 Reactor Cell

-As shown in more detail in Flgs 4.7 and 4 8 the
reactor cell contains the 575-Mw(t) reactor, fuel salt
circulating pump, fuel salt heat exchanger, blanket salt
circulating pump, blanket salt heat exchanger, and the

interconnecting salt piping. The cell has circular ends of
12 ft radius and is about 24 ft wide by 40 ft long by
about 63 ft deep, including the 8-ft-thick roof plugs.

In this design version the major components in the
reactor cell are supported on columns, or pedestals,
which penetrate the floor of the cell. The columns rest
on vibration dampers which are supported on footings
beneath the cell floor structure. The degree of pro-

_tection ‘against seismic disturbances has not been

analyzed. (Subsequent design concepts for a single-fluid
MSBR adopted an overhead support system.): Differ-
ential expansions in the piping and equipment are
partially absorbed by the flexibility of the supports.
Figure 4.7 shows the single 18-in.-diam pedestal for the
reactor vessel hinged at the bottom to reduce the
stresses in the fuel salt piping. Calculations made on the
basis of -a fixed joint, however, gave stresses within
allowable limits. Because of the high ambient tempera-
ture in the reactor cell, the support structure would
probably be fabricated of 304 SS. Bellows joints at the
base of each column would provide the necessary
hermetic seal. '
 

 

 

   

HOT STOBAGE

   

OFF-GAS PROCESS ROOM

REACTOR CELL

22

REACTOR CELL

  

     
        

ORNL-DWG 68-30A

STEAM CELL

AND STORAGE
TANKS CELL

M~STEAM CELL

Fig. 4.8. Plan View of Steam Generator and Drain Tank Cells.

The reactor cell atmosphere will be an inert gas,
probably nitrogen. Since the interior of the cell will
operate at about 1100°F, the cell walls must provide
thermal insulation and gamma shielding to prevent
overheating of the 8-ft thickness of concrete in the
biological shielding. Blanket-type insulation about 6 in.
thick will be used, protected on the inside of the cell by
a thin stainless steel liner which will also serve as a
radiant heat reflector. The construction is shown in Fig.
49. A carbon steel membrane on the outside of the
thermal insulation provides a sealed structure. The
space between this membrane and a surrounding 3-n.-
thick carbon steel thermal shield is also sealed and
continuously pumped down and monitored for leakage
through the inner shell. A second 3-in. carbon steel
plate is separated from the inner plate by a 3-in.-wide
~ air space through which cooling air is circulated. At an
air velocity of about 50 fps the maximum estimated
temperature of the concrete is less than 200°F.!?

The electric heaters for the cells are Inconel pipes
welded together at one end to form a hairpin. Lavite

 

Yy, k. Crawley and J. R, Ros,e, Investigations of One
Concept of a Thermal Shield for the Room Housing a
Molten-Salt Breeder Reactor, ORNL-TM-2029 (November
1967). -

washers separate and support the pipes in the thimbles
in which each unit is inserted. These thimbles are
installed in the permanent portions of the cell roof
structure. With this arrangemeént individual heaters can
be disconnected and removed in event of failure.
Heaters of this type have proved reliable as reactor
vessel heaters in the MSRE.

The reactor cell roof plugs would mcorporate the
same general design features as the walls. Figure 4.10
shows the double barrier at the top of the reactor cell,
at the thimbles for the electric heaters, and also
indicates how the cooling air can be introduced into the
removable roof plugs. Figure 4.9 shows how the double
barrier sealing membranes would be arranged at the
bottom corners of the reactor cell and at the pedestal

supports to permit relative movement. The total heat
‘loss from the reactor cell has been estimated to be

about 2 Mw(t).

The design pressure for the reactor cell is about 50

- psia. In considering the integrity of the cell it should be

noted that no water is normally present which could
accidentally mix with the hot fuel or blanket salt to
cause a pressure buildup through vaporization. To
prevent ‘accidental entry of steam into the cell via the
coolant salt circuit, rupture disks are provided on the

STEAM
PIPING
"

o

")

4]

 

COOLING CHANNEL

 
  

 
 

HEAT ‘REFLECTOR

DOUBLE CONTAINMENT

ORNL-DWG 67-10644A

  
  

EQUIPMENT SUPPORT

K LA

 
 
 

}},,Z//////é/

N

    

    

 

 

 

 

2 E\\>\\\\\\\}\E\\\\\\

 

 

Y

 

   
  

THERMAL SHIELD

024 6810
INCHES

/I’IIIIIIIIIIIIIIII
N

   

 

 

. VIBRATION
« - ISOLATOR

' Fig. 4.9, Cell Wall Construction at Supports.

secondary system which would discharge the coolant
salt into the steam generator cell if there were a
pressure buildup in the system due to a tube failure in

_the steam generator. The rupture disk ratings would be

well below the collapsing pressure of the tubing in the
primary heat exchanger, but even in the highly unlikely
event ‘of tube collapse and shell rupture, sufficient
escape of vapor to cause a significant rise in the reactor
cell pressure does not necessarily follow.

.

4.3.4 Coolant Cell .

Each of these four cells confains a coolant salt
circulating pump, four boilersuperheater units,- two
reheater units, and associated salt and steam piping. The
cells are approximately 24 by 45 ft and about 34 ft
deep, including the roof plugs.

The construction is similar to that used in the reactor
cells in that the cells must be heated and sealed. The

 
 

24

ORNL-DWG 67-10637A
_COOLING SUPPLY
P ,

 

  

 

o hrirreirr b e -
EE LT

 

 

 

  

e e ]
Leleitrs oo 7
T

 

   

 

 

  

1

REMOVABLE

 

  

 

ROOF SLABS

 

 

 

 

 

 

 

 

 

 

 

 
 

e

s :

 

 

 

 

 

 

 

 

 

 

 

 

 

AP

< Lo vd ol L L L

 

 

 

 

 

 

 

 

 

   

 

 

 

COOLING CHA NEL)
LING CHANNEL” DOUBLE CONTAINMENT

 

., é\\\\\\\\\\\\\\\\\\\\i

  
  
 

— CELL HEATER THIMBLE

02 46 810
Letrryreedd
INCHES

Fig. 4.10. Cell Wall Construction at Roof Plugs.

radioactivity, however, is only that induced into the
coolant salt, so there is no need for the steel radiation
shield to protect the concrete. Thermal insulation
would be applied in the same thickness and about the
same manner as in the reactor cell, and the liner would
form the hermetic seal. A double barngr is not required
for containment purposes, but an air flow passage must
be provided to carry away the heat passing through the
thermal insulation to prevent the concrete sh1e1d1ng
from gettmg too hot.

Figures 4.6—4 8 illustrate the arrangement of equip-
ment in the coolant cell. As in the reactor cell, all
components are mounted on support columns which
rest on the floor of the cell. The coolant sait piping is
provided with several expansion loops to achieve the
necessary flexibility without the use ‘of expansion
joints. The expansion of the steam lines is absorbed in

. piping loops located outside the cell. Bellows seals are

provided where the various pipes pass through the
coolant cell walls. Analyses of the stresses in piping and
 

-

0

”

 

equipment indicate that all are within the limits allowed

by the codes.

4.3.5 Drain-Tank Cells

The two drain-tank cells are located as shown in Figs.
4.6 and 4.8. A cross section of the cell is shown in Fig.
4.11. Each cell is about 17 by 50 ft with the end
containing the fuel drain tanks about 73 ft deep. The
other end of the cell houses the blanket and coolant salt

tanks and is about 37 ft deep The walls of these cells

are constructed much the same as the reactor cell walls.
Double containment must be provided, and the cells
must be heated to about 1100°F.

In addition to the various salt lines entering the
draintank cells, there are also pipes to provide for
steam cooling of the tanks and for the inert gas used for
pressurizing the tanks to transfer the salt. (Subsequent
studies  have indicated that a natural-convection salt
system may be superior to a steam system for cooling

the drain tanks.) Bellows seals are used where the piping

passes through the oelI walls.

   
    
    

731t Qin, 7

FUEL SALT
DRAIN TANKS

FLUSH SALT—

25

BLANKET SALT -
DRAIN TANKS

4.3 6 Off-Gas Cells

As will be explamed in Sect. 5.5, the helium that
removes the gaseous fission products from the fuel and
all other contaminated gases are routed to an off-gas
cell for filtration, decay of radioactive contaminants,
and other treatment. The two off-gas cells are approxi-
mately 17 ft X 38 ft X 62 ft deep. The wall and roof
construction is similar to the reactor cell in that double
containment is provided. Since salts are not present,
these cells are not heated and thermally insulated.

4.3.7 Salt Processing Cell

The chemical processmg plant for treatment of the
fuel and blanket salts is contained in a single cell. As

"described in Sect. 4.5, the plant serves all four of the

reactor modules. The cell has a T shape, one leg being
about 10 ft X 34 ft X 62 ft deep and the other about
12 ft X 88 ft X 62 ft deep.

Double containment is required, but since some
pieces of equipment need to be heated and others

ORNL-OWG 68-32A

—_—

STEAM
PIPING

37 ft 10in.

 

 
 

COOLANT SALT
DRAIN TANKS

VALVE {TYP)

Fig. 4.11, Cross Sectional Elevation of Drain Tank Cell,

 
 

cooled, the ambient temperature of the cell is relatively
low. The pipes and vessels would be heated or cooled
individually as required. Biological shielding is needed
because of the high level of radioactivity.

4.3.8 Instrumentation Cell

A centrally located cell is provided for the piping,
junction - boxes, controls, etc., associated with the
instrumentation of the reactor complex. This 10- by

80-ft cell would be operated at normal ambient

temperatures and is not a containment area.

43.9 “Hot” Storage Cell (for Reactors)

A cylindrical cell 20 ft in diameter by 62 ft deep is
- provided at each reactor module for storage of the
reactor vessels and spent graphite cores until most of
the radioactivity decays and they can be processed for
d:sposal The cells are hermetlcally sealed.

4.3.10 “Hot” Cells

A series of small cells, possibly 8 by 8 ft,are shown in
Fig. 4.6 to indicate that cubicles equipped with
remotely operated manipulators and other equipment
will be needed for repair and inspection operations.

4.3.11 Control Rooms, Offices, Shops, etc.

26

appreciable technical difficulties and could be con-
sidered in future steam-system optimization studies.
Supercritical pressure was selected for the steam cycle
because it offered better cycle efficiency, followed an
established trend in the steam power industry,2® and
provided an opportumty for heating the feedwater to
700°F.

The cycle provides for mixing prime steam with the
feedwater to raise the temperature to the inlets of the
steam generators to avoid local freezing of the coolant
salt or excessive temperature gradients in tubing walls.
Future development may show that lower feedwater
temperatures can be used. In this case, sufficiently high
feedwater temperatures possibly could be attained
through additional stages of regenerative feedwater
heating. ORNL4037% and MSR-66-18%! discussed an
alternative cycle in which S80°F feedwater is supplied
to the steam generators and 552°F “cold” reheat steam
is sent to the reheaters. All the feedwater heating would
be accomplished by use of extraction steam from the
turbine. The flow through the steam generator would
be reduced to about 7.5 X 10° Ib/hr, and the boiler
feed booster pumps would not be required. There
would be a saving in the cost of equipment and an
increase in the net overall thermal effimency from

about 449 to 45.4%.

As indicated in Figs. 4.2 and 4.4, space has been

allowed for control rooms, offices, laboratories, shops,
storage, etc.

4.4 Turbine Plant

4.4.1 General

A preliminary study of the MSBR turbine plant was
included in ORNL-3996.! This work has not yet been
extended in any more detail for subsequent MSBR
conceptual design studies.

The steam supplied to the turbine from the steam

‘generator cells would not be radioactive, and no
reasonable accident situation can be conceived where
contamination could enter the steam-circulating system.
The turbine plant is thus conventional with regard to
design, maintenance, and operational procedures.

The relatively high salt temperatures which are
available make it possible to generate 1000°F steam and
to reheat to 1000°F. The upper limit on the steam
temperature was chosen more on the basis of current
steam-system operating practice than on specific limita-
tions of the salt systems. Double reheat would offer no

4.4.2 Turbine Plant Flowsheet |

The turbine plant flowsheet is shown in Fig. 4.12, and
pertinent data are listed in Table 4.1. The flowsheet is
not represented to be the optimum one but rather is
one that appears to be operable and one upon which
preliminary cost estimates can be reasonably based.

Steam is delivered to the turbine throttle at 3500 psia
and 1000°F. After expansion to 600 psia and about
550°F in the high-pressure turbine, it is preheated to
about 650°F, then reheated to 1000°F before returning
to the intermediate-pressure turbines at about 540 psia.
After expansion to about 170 psia, the steam crosses to
the two double-flow low-pressure turbines, where it
expands to about 1.5 in. Hg abs before entering the

- water-cooled condensers. The gross ‘generated output is

about 1035 Mw(e)

 

20p oy C. Robertson, Supercritical Versus Subcritical Steam
Conditions for 1000-Mw(e) and Larger Steam Turbine-
Generator Units, MSR-68-67 (Aprl 24, 1968) (internal corre-
spondence).

zlRoy C. Robertson, MSBR Steam System Per:formance
Calculations, MSR-68-18 (July 5, 1966) (internal corre-
spondence). ‘
 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

      
     

 

 

 

 

 

 

 

 

 

     

 

 

 

 

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   
   
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    
 
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

27
ORNL-DWG 69-6869A
- 5,134,190 # —— e ——
540P-000°F~1518.5 h f -ll
~ 7,152,180 # _
10,067,550 # 1 —_ —_ 152,18 r-= —e soa#_! o _____I
§~ 3600P-1000°F-1424.0h | , | 3515 FI000 F 1424 n i r
,,j | l & . 1,073 # | '
| glg : [t ————— ——=1 | HP TURBINE _\_{_IiE_BINE GENERATOR "A" |
| SIT | I S P GROSS | CONDENSER "A"
I . ol # |l|25617h g ¥ 1990 [ 1376.4n 13‘|fa.1lhl OUTPUT i 1.8in. Hg obs 323.390 4%
| & " = T 527.2 Mw ,
« MR o o# e HE xe * * er
| | 18 9 g ah T wmeew gl gla T Tad LA osotF-TIOn
ssae7,700% | g.e84,100% . | b -Lw'—f-—[———| ______ S o4 S | , Y
1125°F S M25%F Ly & i i | I~ - I
—1280 5,134,190 # 1 i !
REHEAT STEAM —_ 0 g ! i | || |
BOILER- R — PREHEATER . T so0P-55i7eF-iz56.Th T | lflr . i ! L] i | #
SUPERHEATER 100.5 Mw (1) | | Lot )2 | | ¢
293.5 Mw(t) | i I o
1931.5 Mw(1) | : | L ' | i I $|=
s o o e o e s e s e s I —————p—————— | gle
COOLANT SALT —f “COOUANT SALT 570 P~ 650°F ~1323.5h I % o _|r_ T '___} | | | g
850°F _ 850°F re-— | | bmm e ———— oSl
| , | | i . .
e . m—— —— — s e " *
l : i ! | f 240,957 # ' i E I l
s B ' i | | “lr2.496.400% | 243,990 # | == | conpensare
o ; i — o — 2 ot e |
S I | I r = OCO3R r | BOOSTER
° 11 | | e I- .1| : I t- ‘| Pumes
d E | SSR | oy te Turaie sy VY LP ITURBINE B'| GENERATOR'®" | — _
& ol | pemem—————— | ) b L CONDENSER "B I
S 29153708 I ?I"-'-’ : |—-l e ool r |I SRoss | thin. Hg obs I
I ° ]! o <
é '  300P-86RTF 3078 = E{i P I 1 § %5 P i 507.7 Mw : 9N.7°F | 323,390 %
8 B e 1 =l e T ; 59.7h 103°F-71.0h |
2 | | T1g - S {tliél§|3|§=5|c‘z:==|l“illl= ! - | .
T 1 e L e = o - A
| bhor LR EEEE RN l
| ‘ | @ ‘ HOTWELL
_724,520% l_$ 1 i 827,220 % || | 2l ol gl 4 *l &) +l gl I I : | ! : | | - ) BUMPS "B"
I r—— T T T T T || | 2] mf &) 5| m| gl = i [ L I | ]
, I { by { LD jals Ry 9al | pLLbii ] 20w ) |
BF WATER I I by | Pl 1gldgrgi & 8 gl | rryryyr 1 ' i
PRESSURE E‘] A | | | | P L1y i i _[__I_J._...._|_.._|_'_._|.. __l____l i
800STER \J)| ! ! 427,490 % | b I | v i qliiiy 2034794 -
Lo . ] | L T T AT e 0 e T
5 4.6 Mw(e) | | I bt | -l L — 1 —wesr 1= 1=—..1 |
| | | | i THuas ]
| I | | : F gy ke oaonT 0 | | I |
3475P-605°F F——1q k- e | I - R | b -
766.4h l |\ | ,ep s " ! b - =5 ) : 6-B : 7-8 : i fs-8 I—Efa
- - - 4,915,760 # |_L| [_Ll 1'.TA 1.
" —T— 3576090* A— Arte | : - T TV N Y s
‘ = : = A I : i I DC | r--'"'1| = : I | I
: : { | : ! = | DEAERATOR | | | - : : |
‘ | | I VAA EVAPORATOR
b — I L | | I | Lt o
i I i L | “ | \M
I | [ J ‘ e I -1 STORAGE | '
—_ —— B , : )
I o o , . 11 ! | 145,760%) | I I |
I 550.9°F. ‘;A - 2-A . I | . o ] L L L
e - iy y - —
~ 4807°F 428.1°F [ : :372?’:2;3:3“ 362.5°F~ {1 4-A ! T sk 7er T saseer -
475.7°F- 423.45F- I TR . 2. 497,680 # 8-A ]
MIXING TEE o o4 3 h i ! | . = ,447,88 A y y
- : ~— 234 1°F- 173.4°F 133.6°F~ 98.3°F - 930°F-
I ! 329.5h 7,152,180 % 2024h 1413 h 106.5h €63 h
l : . _'_ 13,610 - 4 144, 3 6- 610h
——)——1 -
- |
48?_;";?2: S 43%7fiff2:"’2h ‘ : : : Y | 3.576000% 142,620 220{650 # | 323,390+
' ’ ’ T b0y : 4°F~1513h 148.6/F-HE5h I 143.6°F-11t.5h
&1t r 183.4°F 1513 8.6IF-16.5 | 3.6°F-1IL.5
gl i & ][R TUR BF PUMP | F—" G
ol . “g" | gt
St | | 1
LEGEND Uy |
Ly outpur | | ' {
:TA;AE: g{ l L 14.7 Mw { I } SUMMARY OF PERFORMANCE AT RATED LOAD
T < =i 1 o I
—--— COOLANT SALT - M A I ! i NET OUTPUT 1000 Mw (e}
#  Ib/hr A T | ] ELEC. POWER FOR AUX. 25.7 Mw (e)
P Ib/in2 abs ¢| L__|_|__I__l__.___ ________ | ol | BF PRESSURE~BOOSTER|PUMPS 9.2 Mw(e)
h Br/b I ! I | | GROSS GEN. OUTPUT 1034.9 Mw (e)
. - - 34,040+ | BF PUMPS 29.4 Mw
Ll —————=}- _ _olttg9.0n | REACTOR HEAT INPUT 2225 Mw(1)
—— r b ssaziom GROSS EFF. OF CYCLE a78%
CONDITIONS FOR "8" EQUIPMENT -~ |~ bE=———rf———————- - T atan NET EFF. OF CYCLE 44.9%
- {EXCEPT TUR.-GEN.) SAME AS FOR"A" e 7,152,180 % ¥ ! NET HEAT RATE 7,601 Biu/kwhr
‘ ~3800P=366.1°F-343.3h
) Fig. 4.12. Steam System Flowsheet for 1000-Mw(e) MSBR Power Station,
-

 

 

 

 

 
 

28

 

Table 4.1. MSBR Steam-Power System Design and Performanoe Data
with 700°F Feedwaterd

 

General performance
Total reactor power, Mw
Net electrical output, Mw
Gross electrical generation, Mw
Station auxiliary load, Mw
Boiler feedwater pressure-booster pump load, Mw

Boiler feedwater pump steam-turbine power output, Mw (mechaniEM)

Flow to turbine throttle, Ib/hr
Flow from superheater, Ib/hr :
Gross efficiency, % (1034.9 + 29. 3)/2225
Gross heat rate, Btu/kwhr
" Net efficiency, % .-
~Net heat rate, Btu/kwhr
Boiler-superheaters .
Number of units
Total duty, Mw(th)
Total steam capacity, Ib/hr o
Temperature of inlet feedwater, F
Enthalpy of inlet feedwater, Btu/lb
Pressure of inlet feedwater, psia’
Temperature of outlet steam, °F
Pressure of outlet steam, psia
Enthalpy of outlet steam, Btu/lb op
Temperature of inlet coolant salt, F
Temperature of outlet coolant salt, °F
~ Average specific heat of coolant salt, Btu Ib
Total coolant salt flow
‘1b/hr
cfs
gpm ‘
Coolant salt pressure drop, inlet to outlet, psi
Steam reheaters

Number of units
Total duty, Mw(th)
Total steam capacity, Ib/hr
Temperature of inlet steam, “F
Pressure of inlet steam, psia
Enthalpy of inlet steam, Btullb
Temperature of outlet steam, °F
Pressure of outlet steam, psia
. Enthalpy of outlet steam, Btu/Ib
Temperature of inlet coolant salt, F
Temperature of outlet coolant salt, °F
Average specific heat of coolant salt, Btu 1b -t °F 1
. Total coolant salt tlow _
Ib/hr
cfs
. Epm
- Coolant salt pressure drop, inlet to outlet, psi

-1 0

F-l

 

2225

. 1000

1034.9
25.7

9.2

29.3

7.15 X 10°
10.1 X 10°
47.8

7136

44.9

7601

16

1932
10.1 X 10°
700

769

3770
1000
~3600
1424

- 1125

850
0.41

130
58,300
~60

8
294

5.13X10%

650
~570
1324 -
1000

‘5517

1518
1125
850 -

. 0.41

8.88 X 10°
19.7
8860

~17

58.5 x 10

 

 

-Reheat-steam preheateré

Number of units

Total duty, Mw(th)

Total heated steam capacity, Ib/hr

Temperature of heated steam, °F
Inlet
Outlet

Pressure of heated steam, psia
Inlet ‘
Outlet

Enthalpy of heated steam, Btu/lb

" Inlet
Outlet

. Total heating steam, Ib/hr

Temperature of heating steam, °F
Inlet :
Outlet

" Pressure of heating steam, psia

Inlet
Outlet

Boiler feedwater pumps

Number of units
Centrifugal pump
Number of stages
Feedwater flow rate, total, Ib/hr
Required capacity, gpm
Head, approximate, ft
Speed, rpm
Water inlet temperature, °F
Water inlet enthalpy, Btu/lb
Water inlet specific volume, £t3/1b
Steam-turbine drive
Power required at rated flow, Mw (each)
Power, nominal hp (each)
Throttle steam conditions, psial° F
Throttle flow, Ib/hr (each)
Exhaust pressure, approximate, psia
Number of stages
Number of extraction points

Boiler feedwater pressure-booster pumps

Number of units

Centrifugal pump
Feedwater flow rate, total, 1b/hr
Required capacity, gpm (each)

. Head, approximate, ft o

Water inlet temperature, F
Water inlet pressure, psia
Water inlet specific volumt'3 £t Jib
Water outlet temperature, F

Electric motor drive
Power required at rated flow, Mw(e) (each)
Power, nominal hp (each)

2

8-
100

5.13 X 108

552
650

595
590

1257
1324
2.92 X 10°

1000
869

3600
3544

6
7.15 x 10°
8060
9380
5000

358

330
~0,0181

14.7
20,000
1070/700
414,000
77

8

3

2
10.1 x 10°

9500
1413

. 695
- ~3500
- ~0,0302

~700

4.6
6150

 

4Taken from ORNL-3996 (zef, 1).

 

 
 

 

)}

at

o

" The two feedwater pumps are driven by separate
turbines using steam at about 1100 psia taken from an
extraction point on. the high-pressure turbine. (The
pump turbines can also be driven by prime steam if the

need arises.) The capacity of each pump is about 8000

gpm and the nominal power requirement!? 20,000 hp
each. Eight stages of regenerative feedwater heating are
used, including the deaerator, employing steam ex-
tracted from the high- and low-pressure turbines and
from the feedwater pump turbines. Full-flow deminer-

- alizers will maintain the feedwater purity to within a
few parts per billion. Feedwater enters the steam

generators at 700°F.

29

The steam system is conventional in almost every .

respect except for preheating of the reheat steam and
the heating of the feedwater to 700°F before it enters
the steam generators. As mentioned above, the steam to
be reheated leaves the high-pressure turbine exhaust at

600 psia and about 550°F. It is then heated on the shell -

side of two preheaters by prime steam inside the tubes.
The reheat steam, now at about 650°F and 570 psia,

enters the reheaters, where it is raised to 1000°F by

 counterflow with the coolant salt. The reheated steam

retums to the intermediate-pressure turbme at 1000°F
and 540 psia.

The throttle-pressure heating steam leaving the tubes

.f the preheater mentioned above, now at about 866°F,
s directly mixed with the feedwater leaving the top
extraction heater at about 550°F and 3475 psia. Since
both streams are at supercritical pressure, the mixing

_can be accomplished simply. The resulting 695°F

mixture is then raised to 3800 psia and heated an
additional 5°F by two boiler feedwater pressure-booster
pumps operating in parallel. These 9500-gpm pumps are
shown on the flowsheet as driven by electric motors
(about 6000 hp each) but in an optimized system could
very well be steam-turbme dnven '

443 Layout of Turbine Plant

The relatively high efficiency of the turbine plant and
use of a 3600-rpm turbine-generator make the space
requirements for the turbine plant less than for the
turbine-generator in a water reactor plant of the same
capacity. The layout of the plant, indicated in Fig. 4 4,

is substantially the same as for a conventional station. '

The feedwater heaters, pumps, water treatment equip-
ment, etc., would be located on several floor levels of a

‘building bay provided for this purpose. Use of a
tandem-compounded turbine-generator rather than a
cross-compounded unit would not require slgmficant

alterations to the layout shown.

4.5 Salt Processing Plant

4.5.1 General

A major attractive feature of the two-fluid MSBR is
the relative ease with which the fuel and blanket salts
can be processed to remove fission products, recover
the bred product, and add new fuel. For the reactor to
be a high-performance breeder, the processing must
take place on a fairly rapid cycle and with low holdup
of 233U in the processing equipment. A closely knit
complex of MSBR power stations might make use of a
central processing facility, but the MSBR concept
described here assumes that the processing plant is part
of the 1000-Mw(e) station and serves only the four
reactor modules. _ _

Small side streams of salt are taken from the fuel and
blanket salt circulating systems for processing in an
adjacent cell. A relatively small space is needed for the
processing equipment. A cell having one space about 10

ft X 34 ft X 62 ft deep and another 12 ft X 88 ft X 62

ft deep is provided in Fig. 4.6.

Many of the station facilities such as offices shops,
laboratories, electrical and water services, waste dis-
posal, data logging and analysis equipment, etc., are
shared by the reactor and salt processing plants.

4.5.2 Fuel Salt Processing
Almost all the fissions occur in the fuel salt,-and the

| objective of the fuel salt processing is to keep the

fission product concentrations at a low enough level for
the neutron losses to be acceptably low. This must be
accomplished economically and with low losses of
233y and LiF-BeF, carrier salt. The gaseous fission
products, krypton and xenon, are removed con-
tinuously from the circulating fuel in the reactor as
described in Sect. 4.3.1. In the processing plant the
fluoride volatility process and vacuum distillation are
used to separate the ?33U and the carrier salt from
most of the remaining fission products. Discard of a
small amount of carrier salt is required to remove
fission products that distill with the lithium and
beryllium fluorides.

An overall flowsheet for the salt processing plant is
shown in Fig. 4.13. The fuel salt is drawn semicon-

- tinuously from the circulating systems of the four

reactor modules at a combined rate of about 24 ft3 /day
(corresponding to a 60-day cycle for a reactor with an
average power density of 20 kw/liter) and is collected in
a holdup tank in the salt processing cell. Since the
reactants used in the. processing are not damaged by
irradiation, it is not necessary to provide decay time for .

 

 
 

 

MAKEUP 233UF,

30

ORNL- DWG 67-3654RA

- .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

     
 

 

 

 

 

 

 

 

 

 

 

F, TO FpT0
RECYCLE RECYCLE
| b
UFg +Fy SORPTION | COLD TRAP UFg+ F, +FP | SORPTION | COLD TRAP
N L - . -l
NaF " 'NaF ) o _
' : '
: : .| 233u propucT -
WASTE ~45  WASTE
" NaF : g/day /recctor NaF +FP '
'CARRIER LiF-BeF,
DISCARD | | MAKEUP
LiF + BeF, ‘ '"
~ 170 ft3/day/ reactor 61t3/day /reactor DIST'L'.LAYE_' HE
BLANKET ' - T
Pg DECAY VACUUM : UF+UE, -
FLUORINATI : , e~ Ufy
- FLUORINATION | pemovaL / \ cooL F'TUOR,'NAHON STILL REDUCTION
— -—} CORE o~ - 1 ~ 600°C
=2 | 500-550°C ~1000°C Hp—
_f days ~1mmHg 7 A
LiF-ThF, OR : ;
F2 LiF-BeFo-ThF, | FZJ UF,
- LiF-BeF,-UF,—FP
\ \
WASTE ‘ FISSION PRODUCT
30-year NO. REACTORS/STATION =4 _ ACCUMULATOR
DISCARD POWER / REACTOR = 556 Mw (t) ~ 400ft3
OF SALT FUEL VOL./ REACTOR = 355113 ~ 6 Mw _
BLANKET VOL./REACTOR = 520 ft3 .
U INVENTORY 7 REACTOR & 315 kg LiF-BeF,~UF, RECYCLE FUEL _ Y

 

 

 

 

 

 

 

 

 

 

 

Fig. 4.13, Processing Diagram for Two-Fluid MSBR.

this reason, but up to 24 hr decay may be required for
the fission product heating to be reduced to a level that
will allow proper control of the temperature in the
fluorination process. Removal of decay heat is a
principal design consideration for the holdup tank and
for other equipment in the fuel salt processing plant.
Salt is drawn from the holdup tank into the process-
ing equipment continuously. Some of the processing
operations are of the batch type, but continuous flow is
achieved through the use of parallel flow. paths and
alternate sets of equipment. First, the salt flows to the
top of a fluorination column, where it is contacted by a
stream of fluorine flowing countercurrent to the salt.
The temperature is controlled at about 550°C. The
uranium in the salt reacts with the fluorine to produce
volatile UF4, which is carried overhead by an excess
flow of fluorine. The uranium removal- efficiency is
about 99.9%. Two types of continuous fluorinators are
promising for this application: a falling-drop type,

described by Mailen and Cathers,2? and one in which
the wall of the fluorinator is.cooled to produce a
% -in.thick layer of frozen salt to protect the metal
from corrosion. |

Volatile fluorides of ruthenium, moblum, molybde-
num, technetium, and tellurium are swept out with the
UF,. Fission product iodine and bromine are also
present in the gas. The carrier salt, barren of uranium
but containing most of the fission products, flows out
the bottom of the column for subsequent punficatlon
in a vacuum still.

The UF is separated from the other volatile fluorides
in a series of sorption and desorption operations. The
gas is first passed over pellets of sodium fluoride at

 

22y. C, Mailen and G. 1. Cathers, Fluorination of Falling
Droplets of Molten Fluoride Salt as a Means of Recovermg
Uranium and Plutonium, ORNL-4224 (November 1968).

o
L L]

”

o

L]

»

 

about 400°C, where the fluorides of niobium, ruthe-
nium, and tellurium are irreversibly removed. When the
bed becomes loaded with fission products, it is dis-
charged to waste and refilled with fresh pellets. The exit
gas then flows over a bed of sodium fluoride pellets
maintained at about 100°C, where the UF¢- and MoF,
are sorbed. When this low-temperature bed becomes
saturated with UF, it is taken off stream and the
temperature is raised slowly to about 150°C to selec-
tively desorb the MoF4. The temperature is then
increased further to drive off the UF4, which is

“collected in cold traps at —40 to —60°C. When a cold

trap is loaded with UFg, it is warmed to the triple point
(90°C, 46 psia), drained to the reduction unit for
reconstitution of the fuel salt; and then recycled to the

31

A filter after the reduction unit removes metallic
precipitates from the fuel salt before it is retumed to
the reactor modules.

Experience with batch processing by the fluonde
volatility method dates back to 1954 and includes all
phases of laboratory and development work and suc-

" cessful operation of a pilot plant.?? The process was

reactor. The fluorine carrier gas leaving the low-

temperature sodium fluoride bed is recycled to the
fluorinator.’ Radioactive technetium, iodine, and bro-

mine remaining in the recycled gas decay somewhat,

but about 10% purge and makeup with fresh fluorine is
required to keep the concentrations of these gases
within the desired limits.

The carrier salt, on leé\nng the bottom of the.

fluorinator, enters a vacuum still that is operated at
about 1 torr .and 1000°C. Most of the beryllium
fluoride and lithium fluoride distill, leaving behind the
rare-<arth fluorides, which would have been the princi-
pal neutron poisons in the reactor. The bottom liquid is
recycled through the still and a decay tank as necessary
to control the heating by fission products. The decay
tank has a volume of about 400 ft3, which is judged

sufficient to collect the fission products over the

30-year life of the plant. At the end of this time, the
lithium fluoride and beryllium fluoride in this waste can
be recovered, and the fission products can be packaged
for permanent disposal.

The distilled lithium and berylhum fluorides contain

small amounts of cesium fluoride and rubidium fluoride

and some zirconium fluoride. A small fraction (no more

demonstrated on a large scale in recovering 233U from

the fuel salt in the MSRE. The principles of continuous
fluotination have been demonstrated in the laboratory.
Separation of lithium and beryllium fluorides from
rare-arth fission products by vacuum distillation has
been investigated in the laboratory and has been
demonstrated in an engineeringscale unit by distilling
about half a cubic foot of salt from the MSRE.2* The
reconstitution of fuel salt by hydrogen reduction of
UF, in carrier salt has been demonstrated in the
laboratory.

4.5. 3 Blanket Salt Processmg

The 233U in a two-fluid breeder is produced by the
reaction i

232m+n1>233Th-23_€rfi'n—>233PaWBW“3U '

All the 233U is produced in the blanket salt. A major
objective of the blanket salt processing is to recover the
233y about as rapidly as.it is produced in order to
make it available for addition to the fuel salt to
compensate for burnup. Rapid processing reduces the
inventory of 233U in the plant and the amount of
fissioning that occurs in the blanket salt. The latter is.
important because thorium is difficult to separate from
the rare-earth fission products except by aqueous

. processes, and accumulation of fission products in the

than 5%) of the carrier salt is discarded to purge this

- poison. The very small amount of UF, in the carrier

salt that enters the still 1s also partially volatfllzed and
recovered.

blanket salt would adversely affect the breedmg per-
formance.
- The major ob]ectlve can be ach1eved by processing the A

blanket salt to remove 233U alone or to remove 2*°Pa

- The lithium fluonde—-berylhum fluonde dxstlllate and

the UF4 from the cold traps are added continuously in
the proper proportions to a reducer, where the fuel salt
is reconstituted for return to the reactor modules. The
UF, is dissolved in the salt at about 600°C and reduced

- to UF4 by addition of hydrogen and discharge of -

hydrogen fluoride. The conditions in the reducing
column precipitate nickel and iron that are present as
fluorides due to corrosion of the processing equipment.

and 233U, Removal of 232U alone can be accomplished
by the proven fluoride volatility process, and this is the
method that was proposed for the two-fluid MSBR in
ORNL-3996.! This choice, however, places certain

 

~ ?3W. L. Carter and M. E. Whatley, Fuel and Blanket
_Processmg Development - for Molten Salt Breeder Reactors,
ORNL-TM-1852 (June 1967).

2%W. L. Carter, R. B, Lindauer, and L. E. McNeese, Design of
an Engineering-Scale, Vacuum Distillation Experiment for
Molten-Salt Reactor Fuel, ORNL-TM-2213 (November 1968).

 
 

 

 

 

32

restrictions on the design of a breeder reactor. The
volume of blanket salt in the low-flux region of the

 reactor blanket, or in tanks outside the reactor vessel,

must be large enough so that the average thermal

~neutron flux seen by the 233Pa is about 10'3

neutrons/cm? or less, if the loss by neutron absorption
to form 224Pa is to be kept below 0.5%. The 233U in
the blanket salt must be removed on about a 20-day
cycle in order to keep the fissioning to a very low rate.

Removal of the 233Pa as well as the 233U from the

blanket salt can reduce the volume of blanket salt
required and the thorium inventory by a factor of 2 to
3. Such a process has been conceived, and its basic
principles have been demonstrated in the laboratory.
This is now the preferred method for processing the

several ‘designs for the reactor vessel and the arrange-
ment of the graphite. Two designs finally evolved. One
design is considerably less complicated, but the nuclear
characteristics -are more affected by changes in the
dimensions of the graphite. In the other, radiation-
induced changes in the dimensions of the graphite are
almost fully compensated and would have little effect
on the nuclear characteristics of the reactor. The less

- complicated design is discussed first.

blanket salt for the two-fluid MSBR and is included in

the flowsheet of Fig. 4.13.
The protactinium removal must be on a short cycle to

be fully effective, possibly as rapid as treating the entire.

blanket inventory once every three days, or at a rate of
about 3.6 gpm. The salt is continuously withdrawn
from the blanket circulating system and enters the
bottom of an extraction column to contact a de-
scending stream of liquid bismuth which contains 3000
to 4000 ppm of metallic thorium. The protactinium
and the small amount of uranium in the blanket salt are
reduced to metal by the thorium and dissolve in the
bismuth. The thorium that is oxidized enters the salt.
Thorium is an ideal reductant because the removed
protactinium is replaced by an equivalent amount of
the fertile material. About 96% of the protactinium and

uranium are removed by the process. The protactinium -

and uranium now in the bismuth are extracted into a
second salt mixture; the protactinium is allowed to
decay to uranium, which is released by fluorination to
become the plant product and replaoement fiss1onable
material in the fuel salt.

The bulk of the blanket salt with most of the
protactinium and uranium removed is returned to the
reactor systems, but a small portion is taken off and
discarded to remove accumulated fission products. This
salt is stored until the residual protactinium decays, and

 thie uranium is recovered by fluorination before the salt

is discarded.

s, MAJOR COMPONENTS
5.1 Rmtor

In a molten=salt breeder reactor the 233U fissions in
the fuel salt and heats the salt as it flows through

graphite elements in the reactor vessel. We considered

- A vertical section through the center of the reactor
vessel for one module of a 1000-Mw(e) plant is shown
in Fig. 5.1, and a horizontal section is shown in Fig.
5.2. The dimensions on the drawing are for a reactor
with -an average power density of 20 kw/liter in the
core. Some dimensions for reactor vessels- with other
power densities are shown in Table 5.1. The wessel is
made of Hastelloy N and is almost completely filled
with graphite elements or cells, The central portion of
the reactor core contains the fuel cells. These are
surrounded by several rows of blanket cells. A graphite
reflector is interposed between the blanket and the
vessel wall. Blanket salt fills most of the volume of the
vessel above and below the graphite elements.

‘Fuel salt enters the vessel through a plenum in the
bottom, flows through the fuel cells, and leaves through
a second plenum, also in the bottom of the vessel. The
blanket salt enters the vessel through the side near the
top and flows downward along the wall to cool it The
salt ‘then flows upward through the blanket cells and
through the spaces between blanket cells and between
fuel cells and leaves the vessel through the side near the
top. The channels through the blanket elements and the
spaces between blanket elements are testricted at the
top in order to direct most of the flow through the
spaces between core elements where the heat pro-
duction rates are greatest.

In moltensalt breeder reactors the major changes in
reactivity are made by adjusting the composition of the
fuel salt. Control rods are primarily for making minor
changes in reactivity such as those required for ad-
justing the temperature during operation and for
holding the reactor subcritical at temperatures near the
operating temperature. The design requirements for the
control rods have not been studied in detail. Since one

rod in the center of the core can have sufficient worth

for the easily defined requirements, only one is shown
in the design. It is envisioned as a graphite cylinder
about 4 in. in diameter that would operate in blanket
salt. The rod would move in a graphite sleeve, and
provision would be made for good circulation of
blanket salt through the sleeve. Inserting the rod would
increase, and withdrawing the rod would decrease the
 

i
'
L
|
i
|
!
i
!
|
!
|
!
!
!

a

B

HOLO-DOWN
PLATE

 

 

  

 

CORE 10 ft Oin. DIAM x {3 ft 3in. HIGH
K420 FUEL CELLS AT 5¥%8 HEX

33

14 ft Oin. DIAM

CRNL-DWG 67-406384

 

/CONTROL RCD

 

 

 

 

REACTOR
VESSEL-+

 

REFLECTOR
GRAPHITE

BLANKET
CELLS—

‘FUEL' SALT _
: PLENUMSZ

 

 

\
\
0

 

 

 

 

 

 

    
   

  

 

TO
BLANKET HEAT

EXCHANGER

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

%/L
I
/
v
]

i
{6in. N.P.S

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

19 f G in.
13 £ 3in.
CORE
l
|
1 — |
1T 2 4 2in.
S22 |
\ . 2ttéin _ =
~ N _  }—=—suPPorT LEGS
' T (8)
it

 

 

 

 

“Fig. 5.1, Vertical Section Through Center of Reactor Vessel for One Module of 1000-Mw(e) Plant.

 
o

i

T

| PRESSURE

" VESSEL

| CONTROL
ROD

 

ORNL-DWG 67-10643A

 

 

14 ft Oin. OD—
10 ft O in. DIAM CORE

 

   

  
 

 

 

S R F
AR LT

HOLD-DOWN PLATE
FUEL SPACER SHEETS

BLANKET CELL SPACER TIES

 

                 
    

     

’

    

    

 

— M D@ O

BT BBEGN
/9 K 000000¢ (X 00%&%&.% NG IR0
. ‘%Pw%%bo&mo»%a 5 ,,w%%wf??hfi
R R AN e
RS O]
NG ;

R0
?WOO‘A 0 ,VOS R :
10 X R NG P OO0
W.a.:.““. R
L8l X XRR)

              

O R O e
LR et O K K X el
N % 04000000 SO 000 000%.%0 0?.
,am...v“?w XX I %000 000? (X) Sl
NG Tl (R
RS eereIereiereie LXK (O 1
N pRieleleeraraieraie) ¢ X X (007
e O
, NN M AR N N N N A R PR A XA

BLANKET PLENUM

LA BTN

SRR

 

R

S OO0k

COOLING SKIRT

SR h
A ey ce

 

 

5 [

= o "nw=
nwe 2 42 -~
£ 9 @eg 3
) oX o O o= e
OLuc 3S Swb
AT i -
Y- <z Zat YEe
FA&m:a < o0 <
o™ = 03 U 3
D x oo O~
L o

 

so that the volume fractions

0.134 fuel salt, and 0.064 blanket

?

The fuel cells are arranged in the core on a triangular

spacing of 5% ¢ in. pitch

metal tube that is welded into the fuel outlet plenum.
are 0.802 graphite

Fuel flows in and upward through the annulus between
the concentric tubes and downward and out through

plenum. The inner giaphite tube is a sliding fit over a
-the bore of the inner tube.

salt. The blanket elements are simple cylindrical tubes

5% in. OD by 3% in. ID, also arranged on 5% in.

triangular pitch, This provides volume fractions of 0.58

blanket salt and 0.42 graphite in the blanket region. For

Fig. 5.2. Horizontal Section Through Center of Reactor Vessel.

A sectional drawing of a graphite fuel cell is éhbwn in -
Fig. 5.3. For the reactor with an average power density

of 20 kw/liter, the cell has an outer hexagonal tube 5%
in. across flats with a 22%;,-n.-diam bore. Inside this -

tube is a concentric tube 2% in. OD by 175 in. ID. The

hexagonal section of the element is about 13 ft long;
end sections are reduced in diameter to provide for

blanket regions at the top and bottom of the core. The

outer graphite tube is brazed to a metal piece at the
bottom end, and this piece is welded into the fuel inlet

reactivity. Rapid movement does not appear to be

necessary.

 

 
n

 

35

Table 5.1. Variation of Some Reactor Characteristics with Power Density and Design Lifetime

Blank spaces in table represent data that were not fully developed
since the reference design was taken as the 20-kw/liter case

 

Average core power density, kw/liter
Design lifetime, full-power years?
Power, Mw()4 '
Core diameter, ft
Core height, ft
Core volume, £t
Fraction fuel in core
Fraction blanket in core
Blanket thickness, ft
Fraction salt in blanket volume
Fraction salt in graphite
Number fuel cells in core
Number blanket cells in core
Overall length of fuel cell, ft
Overall length of blanket cell, ft
Reflector thickness, ft
Fuel-salt volumes?
Reactor core, ft3
Plenums and piping, fe3
Heat exchangers and pumps, ft3
Processi:;g plant, ft3
Total, ft
Salt processing cycle times, days
" Fuel stream
Fertile stream
Pa removal stream
Breeding ratio
Fuel yield, % per'year
Fuel cycle cost, mills/kwhr
Fissile inventory, kg A
Fertile inventory, kg
Specific power, Mw(t)/kg :
Average flux, >0.82 Mev, 1012 neutrons cm ™2 sec™!

>0.50 kev, 10'? neutrons cm ™2 sec ™

10 20 40 80
17.2 8.6 43 2.2
556 556 556 556
12 10 8 6.3
18 ' 13.3 10 8
2036 - 1041 503 253
0.098 0134 0.154 0.165
0.058 0.064 0.067 0.06
1.0 1.25 1.25 1.35

0.58

0.42

240

252

15.3

15

0.5

139

37

160

6

355
173 110 77 50
144 110 70 50
1.4 1.1 0.7 0.5
1.05 1.06 1.06 1.05
2,75 4,07 5.01 5.59
0.6 0.5 0.5 0.4
413 315 - 261 220
63,000 54,000 39,000 31,000
1.35 1.77 2.13 : 2,53
1.77 3.33 6.72 13.1

0.50 ' 094 1.90 3.70

 

' @The design lifetime is based on an allowable fluence of 5.1 X 10%2 'nefitxonslcmz (see ref. 15, Sect. 3.4).

bper reactor module,

reactors with core power densities different from 20

kw/liter, the dimensions of the fuel and blanket cells

and their spacings are adjusted to provide the desired

sizes of core and blanket and volume fractions of
materials. | -

In Sect. 34 we indicated that the graphite could be
expected to contract and then expand when irradiated

in the core of an MSBR. The useful life for design

purposes is assumed to be the time for graphite to be
irradiated to a fluence of 5.1 X 10?2 neutrons/cm? (E

> 50 kev).2® With fluence limiting, the design lifetime
of the graphite varies inversely with the damage flux,
which in turn is proportional to the power density in
kilowatts per liter of core volume. By properly varying
the volume fractions of fuel and blanket salt with
position in the core, a ratio of maximum to average
power density of 2 can reasonably be obtained. A core

 

5gee ref. 15, Sect, 3.4, relative to current values for limiting
neutron fluence.

 
 

 

 

ORNL-DWG €7-10646A

 

 

 

 

      
 
  
 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1
- 3in
: /,
1ft 3in. e,
3|I"I // // N
7, / /, / \\\
7 < €& ‘///\ N\
///’?/r‘\ \\\wa/z/\\\\\
rC/Z\flf’/\\\\
VDN 'Z" AN
AR
X . "/’\\\ NN \\:\
NN
=Y NN
N ] NN
7 A SN\
& % ZINN
2\__2 ‘\\\ >
13t 3in.
REACTOR
5.375in. 1% in. 10x2Y%in.0D
22%35in.1D ’
| SIX % -x1%-in. SLOTS
N _
ST '
SH AR 3%in. DIAM
N ZIN ‘
1Qin. N7 AN ———Yin,
N7 AN
W% R
WZ RS
NA  UN
N AN
N A 1IN _
NA U
NIZ ZIN GRAPHITE TO
3 \A jfiv = METAL BRAZE
\\ ( _ 17gin. 0D x43/4in. ID
, : tYyin.0Dx4Ygin. 1D
3in. MIN . —1% 8
N (Y
" FUEL INLET PLENUM

 

 

 

| SS~—— FUELOUTLET PLENUM

. Fig. 5.3. Sectional Drawing of Graphite Fuel Cell.
 

-y

with an average power density of 20 kw/liter would
have a maximum power density of 40 kw/liter, a
maximum damage flux of 1.9 X 10'* neutrons/cm?,
and a design lifetime of 8.6 full-power years, or 10.8
years with an 0.8 plant factor. Table 5.1 shows how
some of the characteristics of a reactor for one module
of a 1000-Mw(e) plant would vary with average power
density and design lifetime.

Under irradiation the isotropic graphite being con-
sidered at the time of these studies would decrease in
volume by 7.5% during the contraction stage and then
would increase in volume by as much as 7.5% ower its

initial volume by the end of its useful life. These

changes in volume correspond to changes in linear
dimensions of +2.5% over the initial dimensions and
create several design problems. The overall lengths of
the graphite fuel cells would change by several inches
during the lifetime of a core and would vary with
location in the reactor. We preferred not to use a

beliows in the fuel salt line to each fuel cell and favored

a minimum number - of graphite-to-metal seals. We
therefore chose to have the fuel enter and leave the
bottom of the fuel cell so that each element would have

“only one metal-to-graphite brazed joint and the graphite

would be free to contract and expand axially, as shown
in Fig. 5.1,

The change in radial dimensions presented a more
difficult problem. Densification of the graphite to
produce a 2.5% reduction in distance across the flats of

the hexagonal tubes would cause the fraction of the

cross section of the core occupied by fuel cells to

“decrease by 5%, and the space occupied by the blanket

salt would increase correspondingly. For the reactor
with an average power density of 20 kwf/liter, the
volume fractions in the core would change from 0.802
to 0.762 for graphite, 0,134 to 0.127 for fuel salt, and
0064 to 0.111 for blanket salt. Changes of equal
magnitude, but opposite in direction, would occur

- during the expansion phase. The rates of change of

dimensions would vary with local power density, so at
no time during the life of a core would the volume

fractions corresponding to the maximum contraction or

expansion exist throughout the core. At the end of life
the graphite at the center of the core would have
reached its maximum volume; graphite in the regions of
average power density would be about at its minimum

- volume, and graphite in the outer fuel cells would be

about halfway into the contraction stage.

Stresses arise in the graphite from dimensional
changes due to gradients in temperature and neutron
flux. A maximum tensile stress estimated to be about
700 psi would occur at an axial position slightly above

37

the center of the core. In subsequent, more detailed

“analyses of graphite elements of similar configuration in

a one-fluid reactor concept, the maximum stress was
calculated to be 500 psi.?® These stresses are all well
below the tensile strength range of 4000 to 5000 psi of
graphites being considered for use in MSBR’s. 42 ¢

No nuclear calculations were completed to show how
the fuel salt and blanket salt compositions would have
to be adjusted to compensate for the change in volume
fractions and how the adjustments would affect the
performance, However, the power-flattening calcula-
tions showed that the power distribution in the core
was quite sensitive to the local volume fraction of
blanket salt. We concluded that a design in which the
volume fraction of blanket salt varied so widely was not
likely to be satisfactory; thus we looked for an
alternative 2

An alternative design for the reactor vessel is shown in
Fig. 5.4. The graphite fuel tube assembly for the core of
this reactor is shown in Fig. 5.5. Blanket cells are

“simply cylindrical tubes of graphite 6% ¢ in. OD by 5

in. ID, each with a metal tube brazed into the upper
end. The reference design.concept described here is’
again for a reactor with an average power density of 20
kw/liter in the core. Basic dimensions of reactors
designed for other power densities are those in Table
5.1,

The primary difference between this design and the
one just described is that in this case the blanket salt in
the core is confined to the annulus between fuel-
containing tubes and the outer tube of graphite
fuel-tube assemblies. The salt in the blanket region is

* confined to the inside of the blanket cells. To accom-

plish this the blanketsalt-containing tubes are con-

- nected to plenums in the top of the reactor vessel and

dip into a pool of blanket salt in the bottom of the
vessel. Helium fills the space between core assemblies
and between blanket assemblies at a pressure that is
controlled to provide the desired level of blanket salt in
the bottom of the reactor vessel. : :
In this design the changes in axial dimensions are ac-
commodated as before. The graphite tubes are fastened

 

26Dnnlap Scott and W. P. Eatherly, “Graphite and Xenon

Behavior and Their Influence on Molten-Salt Reactor Design,” .

Nucl, Appl. Technol. 7(8) (February 1970).

27Results of more recent tests (December 1968) indicate that
some isotropic graphites undergo little change i m volume during
irradiations to at least 2,6 X 10? 22 jeutrons/cm? (E>> 50 kev),
the maximum exposure obtained in the tests. Availability of
such materials in the desired sizes and shapes would ehmmate '
the major objection to this demgn

 
 

 

CORE CELL——__|

~ PRESSURE VESSEL~

BLANKET
PLENUM

"RADIAL .

_ BLANKET OUT

S,

1z

[~

BLANKET CELL 198

REFLECTOR
GRAPHITE———____

T I T I T

T 2T 2T

 

 

 

o i A A 7 A . S S T A A A 3 S S S A N S S &

BLANKET OPERATING

LEVEL\l

r
X

 

 

 

 

 

SO

 

 

 

38 .

CONTROL ROD

ORNL—-DWG 68—29A

 

13 f+ 10 1n. DIAM —

CORE 8t 9in. DIAM
X143 ft 3 In. HIGH

240 FUEL CELLS

AT 6% PITCH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T2

 

7

T YT YT A Tl T T R

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Lz £

 

BLANKET OUT

 

FUEL SALT PLENUMS

 

 

 

 

 

 

19 f+ 2 in.

 

 

 

 

2ft Tin.

 

“Fig. 5.4, Vertical Section Through Center of Alternative Design of Reactor Vessel for One Module of 1000-Mw(e) Plant,
ORNL ~DWG 68-31A

 

BLANKET PLENUM

39

i

 

 

 

2625 5000

                   
 

—1.625

 

   

 

 

 

w
N
d
/ V2 &
S
.///‘//// 2
227 2
w
r =
- 2
w
Q b
M~ = T
< a % &
o e Q
Y . & &
" - o
© o

    

///

\\_

DRESUNNNY

//// // ///////////

 

 

\ /
N

 

 

FUEL QUTLET PLENUM

1% 0D x1% 1D
1% 00 x1' 1D

 

~—— FUEL INLET PLENUM

=

A —

 

 

 

  

       

N \\\\ o
TR 7///////, O //// //7/4. R e S

\///

 

T TS

 

 

 

 

f—— 12

 

Fig. 5.5. Graphite Fuel Tube Assembly for Core\_ll_{e'gion.

 

 

4
1313m
20

 

 

 

18 ft 2in.

DIMENSIONS IN INCHES
EXCEPT WHERE NOTED
 

 

to a metal structure at one end only and are free to
move axially. With blanket salt and fuel salt confined
by graphite tubes in the core region, radiation-induced
changes in the transverse dimensions of the graphite will

produce proportionate changes in volumes of graphite,

40

blanket salt, and fuel salt. The relative volumes of these -

materials would remain about constant, and the only
major changes in fractional volume would occur in the
gas spaces between elements. Although the nuclear
characteristics would vary some with time (the amount
had not been calculated when the work was inter-
rupted), it would be surprising if there were a large or
serious effect. _

In this design the fuel salt enters the reactor vessel
through a plenum in the bottom, flows through the
reentrant tubes of the fuel tube assemblies, and leaves
through a second plenum in the bottom of the vessel.
The blanket salt enters through a plenum in the top of
the vessel and flows downward through the outer
annulus of the fuel tube assemblies and into the pool of
blanket salt in the bottom of the vessel. Two-thirds of
the blanket salt flow goes out through a pipe from the
bottom of the reactor vessel to the suction of the
blanket salt circulation pump. The discharge from this

pump, after passing through the blanket salt heat

exchanger, enters four ejector-type jet pumps operating
in parallel. The suction side of these jets is connected to
the radial blanket plenum in the top of the reactor
vessel. The jets draw blanket salt upward through the
radial blanket cells and discharge the combined flow
into the plenum that supplies the core elements. This
method was chosen for circulating the blanket salt
because it seems to overcome the problems of distri-

buting the flow between the core elements and radial

blanket elements while assuring that the elements will
be kept full of salt.

5.2 Fuel Salt Primary Heat Exchanger

Each reactor module has a fuel salt primary heat
exchanger in which the fission heat in the fuel salt is
transferred to the coolant salt. The exchanger is of the

vertical countercurrent shell-and-tube type with the fuel

salt in the tubes. The impeller and bowl of the fuel salt
circulating pump are an integral part of the top head
assembly of the heat exchanger. The pump will be
discussed separately in Sect. 54.

The general configuration of the exchanger is shown
in Fig. 5.6, and the principal data are given in Table 5.2.
Each exchanger is about 6.5 ft in diameter X 20 ft high

and has an effective surface of 12,230 ft®. All portions

Table 5.2. Fuel Salt Primary Heat Exchanger Data

 

Number required per reactor module 1
Rate of heat transfer, Mw 5.29

Rate of heat transfer, Btufhr 1.80 X 10°
Total surface, ft* 12,230
Shell side
Hot fluid or cold fluid Cold (coolant salt)
Entrance temperature, °F 850
Exit temperature, °F : 1110
Entrance pressure, psi 198 .
Exit pressure, psi 5 164
AP across exchanger, psi - 34
Mass flow rate, Ib/hr 1.68 X 107
Tube side
Hot fluid or cold fluid o Hot (fuel salt)
~ Entrance temperature, F 1300
Exit temperature, °F 1000
Entrance pressure, psi 146
Exit pressure, psi 50
AP across exchanger, psi . 96
Mass flow rate, Ib/hr 1.09 X 107
Velocity in tubes, fps : L~ '
Tube material ' Hastelloy N
" Tube OD, in. 0.375
Tube thickness, in. 0.035
Tube length, tube sheet to
tube sheet, ft :
Inner annulus 15.3
Outer annulus 16.7
Shell material ' Hastelloy N
Shell thickness, in, 1
Shell ID, in. 67
“Tube sheet material Hastelloy N
Tube sheet thickness, in. .
Top outer annulus 1.5
Top inner annulus 2.5
Floating head 3.5
Number of tubes
Inner annulus 4347
Outer annulus 3794
Pitch of inner annulus tubes, in.’ :
Radial 0.600
Circumferential 0.673
Pitch of outer annulus tubes, in. 0.625, triangular
Type of baffle Doughnut
Number of baffles .
Inner annulus 4
Outer annulus _ 10

 

in contact with the fuel and coolant salts are con-
structed of Hastelloy N. The pump tank, which is about

- 6 ft in diameter X 8 ft high, is mounted directly above

the heat exchanger and is part of the pump and heat
exchanger assembly. A S-n, fill-and-drain line connects
the bottom of this tank to the fuel salt drain tanks.
 

 

41

ORNL-DWG 67-10642A

 

  
 
 
 
 
 
    
  
  
  
 
  
    
 
  

o] "'}r__"';____—~—————~——moroa FLANGE
L. j .
23t 7"4"1. u"*-~..“,_—~___./
]

GAS CONN

START-UP LEVEL

6-ft O-in-0D TANK

TO HOLD-UP TANK
HIGH OPERATING LEVEL

291t 3% in. -
, LOW OPERATING LEVEL

AND DUMP

Sin. NPS -

MCLTEN SALT BEARING

2 1%5in.

  

16in. NPS
FUEL TO REACTOR

11t Sin. i6in. NPS

  
     
   
 

11t Sin.
. IMPELLER

TUBES
3732 AT 3in. OD

20ft 2in,

Lo

INNER TUBES
4347 AT 34in.0D

 

 

 

 

 

 

 

 

 

 

 

 

J
‘ ;
15 ft 3in, A : ML LR Bt :
’ .
' ] el 6t Tin. 0D
4 Bbl A A : ..'i"”l'n i !
4 y f
;
: . M ~
2-in 21t tin, /A
DRAIN LINE — _ = —
1 ft in.

 

 

 

 

 

440 NPS (2}—<]——{ L_’_\_I——'zom. NPS

Fig; 5.6. Fuel Salt Primary Heat Excfianger and Pump Assembly for 250-Mw(e) Reactor Module,

 
 

 

 

 

Fuel salt flows from the reactor at 1300°F through
the 16-in. pipe connected directly to the top of the
circulating pump. The pump boosts the pressure from
about 9 psi to approximately 146 psi and discharges the
~ salt downward through 4347 bent tubes to the lower

42

tube sheet. The flow direction then reverses, and the

salt flows upward through 3794 straight tubes in the
outer annulus, or bank, of tubes and leaves the
exchanger at about 1000°F. The tubes in both banks
are % in. OD, and the salt velocity in the tubes averages
about 9 fps. Using a tube sheet at the bottom, rather
than employing U-tubes, provides a plenum for draining
the fuel salt from the exchanger. A loop in the 2-in.
drain line inside the shell provides the necessary

The number of tubes in each of the annular regions
was determined on the basis of the desirable pressure
drop for the fuel salt flow through the tubes and on the
allowable temperature drop across the wall. The heat
transfer coefficient in the inner annulus needed to be
lowered to minimize the temperature gradient through
the wall; the velocity was therefore reduced by using
4347 tubes as compared with 3794 in the outer
bank.2® The length of the tubes was determined largely

on the basis of preliminary calculations which showed

that 15 ft would provide about the desired geometry.
Baffle spacing in the inner annulus was fixed by the

~ distances required for the unrestrained bends in the

flexibility for thermal expansion and movement of the

bottom tube sheet. -

The bent tubes in the inner annulus accommodate the
differential expansion between the inner and outer
banks of tubes. To simplify the bends, the inner tubes

are placed on concentric circles with a constant delta

radius and 2 nearly constant circumferential pitch. A
radial spacing of about 0.6 in. was selected as being the
minimum practical pitch. The tubes in the outer
annulus are located on a triangular pitch of 0.625 in.?®
" The 850°F coolant salt enters at the bottom through
two 14-in. pipes at a pressure of 194 psi and flows
upward through the outer annulus to cool the vessel
outer wall. It then reverses direction and flows down-
ward over the outer bank of tubes in a counterflow
arrangement. At the bottom tube sheet it again reverses
direction and flows upward across the inner bank
of tubes. Doughnut-shaped baffles are used in both
annuli. The salt then leaves through a 20-in. coolant salt
pipe at the center line of the exchanger at about
1111°F and 161 psi. Drain ports, not shown in Fig. 5.6,
allow the coolant salt to be drained from the space
above the lower tube sheet.

The heat transfer and stress correlations used in
conceptual design of the heat exchanger have been
reported by Bettis ef al.2® The properties of the fuel
and coolant salts and of the Hastelloy N used in the
_ calculations are given in Tables 3.1 and 3.4. Comiputer
codes were written to optimize the salt-tosalt and
salt-tosteam- MSBR heat transfer equipment. Except
for some work on the steam generators, howewer, the
~ codes were not fully operational when this equipment
was designed.

 

- 28GE&C Division, Design Analysis Section of ORNL, Design
Study of a Heat Exchange System for One MSBR Concept,
ORNL-TM-1545 (Sept. 1967).

tubes and the maximum allowable temperature drop
across the walls. The spacing in the outer annulus was
selected to give the most efficient use of the shellside
pressure drop. Ten baffles were used in this region.?®

In this conceptual design, individual tubes cannot be
inspected, repaired, or replaced. Reliance is therefore
placed on quality control in the manufacture and
installation to obtain highly reliable units. Should major
difficulties develop in an exchanger, however, it would
be necessary to replace an entire heat-exchanger—pump

. assembly, as discussed in Sect. 5.10. The rotating parts

“through the inner bank of %-n.-OD tubes, to the

of the pump can be replaced with relatively little
difficulty.

As may be noted in Table 3.1, the thermal con-
ductivity of both the fuel and coolant salts is now
known to be substantially less than the values used in
design of the heat exchanger presented here. Use of the
newer values would increase somewhat the amount of
heat transfer surface and the inventory of salt.

5.3 Blanket Salt Primary Heat Exchanger

Each reactor module has a blanket salt heat exchanger
for transferring heat from the blanket salt to the
coolant salt. The exchangers are about 4.7 ft in
diameter X 19 ft high overall and are of the vertical
shell-and-tube type with the blanket salt in the tubes, as
shown in Fig. 5.7. Although smaller, the units are very
similar to the fuel salt heat exchangers and have the
same arrangement of the salt circulating pumps as an
integral part of the top head. Hastelloy N is used for all
portions in contact with the salts.

The blanket salt is cooled from about 1250 to
1150°F in its passage through the exchanger. The flow
is from the reactor, through the pump, downwara

bottom tube sheet, where the flow turns upward

through the outer bank of tubes. Unlike the fuel sait
 

43

ORNL-DWG 67-10629A

- T . L2 MOTOR FLANGE

21t Oin. DIAM~%:——-—$r

= 301t 6¥in. S e

 

   
   

SPOOL

 

4ft Oin.
‘ OPERATING LEVEL

MOLTEN SALT BEARING

PROCESSING
DRAIN LINE

BLANKET TO
REACTOR 8in. NPS

   
  
  
  
   
  
 
 
  
 
  
 
 
  
 
   

8in. NPS

GAS IMPELLER

&)

14§t 1%,in.

19f¢ Oin. .
8ft 4in,

TUBES

GAS SKIRT
822 AT ¥gin. 0D

TUBES

4 £t 8in, ]
834 AT ¥gin. 0D

 

COOLANT 20in. NP5

2t 9%in.

- 44in. NPS

= Fig. 5.7. Blanket Salt Primary Heat Ext:hénggr‘and Pump Assembly for 250-Mw(e) Reactor Modute,

 
 

 

exchanger, straight .tubes are used in both banks. The
pertinent data are given in Table 5.3.

* The coolant salt is circulated in series through the fuel
salt and blanket salt exchangers. The salt leaves the fuel
salt exchanger at.about 1111°F and is heated to about
1150°F in the blanket salt unit, absorbing about 28
Mw(t) of heat per reactor module. The coolant makes
one pass through the shell side, entering through the
20-in.-diam central column, flowing downward between

 

the disk-and-doughnut baffles, and exiting through a

20-in.-diam side nozzle.

Table §.3. Blanket Sait Primary Heat Exchanger Data

 

Btuhr? £t

Number required 4
Rate of heat transfer, Mw 27.8
- Rate of heat transfer, Btu/hr 9.47 X 107
Shell side 7 ' S
Hot fluid or cold fluid Cold (coolant salt)
Entrance temperature, °F 1110
Exit temperature, F 1125
~ Entrance pressure,? psi 138
~ Exit pressure,? psi 129
AP across exchanger,b psi 15
Mass flow rate, Ib/hr 1.68 X 107
Tube side
Hot fluid or cold fluid Hot (blanket sait)
Entrance temperature, °F 1250
Exit temperature, °F 1150
Entrance pressure,? psi 111
Exit pressure,? psi 20
AP across exchanger,” psi 91
Mass flow rate, 1b/hr 43 X 10%
Velocity, fps 10.5
Tube material Hastelloy N
Tube OD, in. 0.375
Tube thickness, in. 0.035
Tube length, tube sheet to tube 8.3
sheet, ft
Shell material Hastelloy N
Shell thickness, in. 0.50
Shell ID, in. 55
Tube sheet material Hastelloy N
Tube sheet thickness, in. 1
Number of tubes
- Inner annulus 834
Outer annulus 822
Pitch of tubes, in. 0.8125, triangular
Total heat transfer area, ft? 1318
Basis for area calculation Tube OD
Type of baffle Disk and doughnut
Number of baffles 4
Baffle spacing, in 19.8
Disk OD, in. 33.6
Doughnut ID, in. 31.8
Overall heat transfer coefficient, U, 1030

 

4Includes pressure due to gravity head.
- bPressure loss due to friction only.

44

Drainage of the blanket salt can be accomplished
through a drain line at the bottom of the tube sheet,

‘not shown in Fig. 5.7. A large pump tank is not

- as a function of the baffle spacing.

required, as in the fuel salt system, since the reactor
blanket volume is filted with salt before circulation is
started. ©

‘Essentially the same heat transfer relatxonshlps were
used for analysis of both the fuel and the blanket salt
exchangers. The number of tubes per pass in the
blanket unit could be established in a straightforward

‘manner, but determination of the baffle spacing and the

tube length that fulfilled both the heat transfer and
pressure-drop requirements became involved. The first
step was to generate data for the outside film resistance
It was next
determined whether the baffle spacing was limited by
the thermal stress in the tube wall or by the allowable
shellside pressure drop. Equations were then developed .
to relate the baffle spacing, the outside film resistance,
and the pressure. drop, as described in ORNL-TM-
154528

5.4 Salt Circulating Pumps
5.4.1 General

The 1000-Mw{e) MSBR conceptual design employs
four fuel salt circulating pumps and four blanket salt
pumps, one of each for each of the four reactor
modules. The pumps are integral with the primary heat
exchangers, as illustrated in Figs. 5.6 and 5.7. There is
also a coolant salt pump located in each of the four
coolant cells. These pumps are somewhat larger but are
similar to the fuel and blanket salt pumps. '

All the pumps are vertical-shaft, sump-type, single-
stage centrifugal units and are driven by electric motors.
The fuel and blanket salt pumps operate at a constant
speed of about 1160 rpm, while the coolant salt pumps
operate at speeds that are variable between about 300
and 1160 rpm. The principal data for the three types of
pumps are given in Table 5.4. ,

As shown in Fig. 4.7, all the pumps have the electric
drive motors located in sealed housings at the operating
floor level. This facilitates access to the motors for
maintenance, and they can be shielded to protect
electric insulation and lubricants from radiation
damage. The motor housing is thus an integral part of

~the containment system and is subject to the same

integrity requirements.
The fuel and blanket salt pumps have long shafts with
the impeller and casing located some 30 ft below the
[3]

&

 

45

Table 5.4. Salt-Circulating Pumps for the

 

 

1000-Mw(e) MSBR
Fuel  Blanket Coolant

Number required 44 44 44
Design temperature, °F 1300 1300 1300
Capacity, gpm 11,000 2000 16,000
Head, ft 150 80 150
Speed, rpm - 1160 1160  300-1160
Specific speed, N, 2830 2150 3400
Net positive suction head required, ft 25 8 32
Impeller input power, hp 990 250 1440
Distance between beanngs ftb 29 29 1.5
Impeller overhang, ft? 2.5 25 . §

 

90ne pump is required foreach of the four modules in the MSBR.
DEstimated from preliminary pump layouts.

drive motor. The upper bearing for the shaft is

. oil-lubricated, but the bearing at the lower end is
lubricated by the pumped salt. In general, a short-shaft .

pump with an overhung impeller and all bearings of the
oil-lubricated type are desirable features since there are
fewer development problems with regard to both the
salt-lubricated bearings and the rotor dynamics. Long
shafts were used in this MSBR two-fluid design concept,
however, because the fuel salt enters and leaves at the
bottom of the reactor vessel, placing the pump casing at
about this same elevation conserves salt inventory, and
we preferred to locate the drive motors outside the
reactor cell,

The number of reactor modules selected for the
two-fluid MSBR design study was influenced by the size

of pump that appeared to be a reasonable extrapolation .

of the MSRE pump size. One salt pump was assumed

per circuit, on the basis of a study made by Grindell -

and Young,?® which found that parallel operation of
pumps in which the liquid level in the pump bowl is

maintained by a gas overpressure could lead to in- -

stability problems.
‘Only preliminary studies were completed on the
conceptual design of the salt pumps. Work had pro-

5.4.2 Fuel Salt Circulating Pump

A design concept for the fuel salt circulating pump is
shown in Fig. 5.8. The oil-lubricated ball bearings and
shaft seal at the top are similar to those which have
performed satisfactorily in the MSRE. The seal consists
of a Graphitar stator bearing against a tool steel rotor.
Lubricating oil is on one side of the seal, and helium gas
is in the shaft annulus on the other side. The gas, in
flowing upward through a labyrinth seal, prevents
movement of lubricating oil vapors downward: and
scavenges oil vapors from the system. A downward flow

of the purge gas is also provided along the shaft to

retard the upward diffusion of salt vapors and fission
product gases.
Some preliminary development was done on the

salt-lubricated bearing for the lower end of the 34-ft.

long pump shaft. One study®® indicated that self-acting
hydrodynamic film lubrication was to be preferred over

-the externally pumped hydrostatic type of film. The

gressed sufficiently to indicate, however, that a careful

study of the effect of pump shaft and casing sizes on -

- relatively high viscosity of the molten salt provides
- good load capacity and hydrodynamic film operation in

the laminar regime. A tilting (pivoted) four-pad type of
self-acting bearing design was selected as being the most
stable.®! The bearing would be constructed of Hastel-

“loy N with a special hard-surface coating. Four of the’

the rotor dynamics would probably be required. Selec-

tion of suitable materials for the salt-lubricated bearing
had just begun when the work was terminated. -

 

29A. G. Grindell and H. C. Young, Two Parallel Pumps

‘coating materials under consideration were: (1) cobalt-

 

3°Féasibi1fty' Study of Rotor-Bearing System Dynamics for a

: 1250-hp Molten-Salt Fuel Pump, MTI-68TR9, Mechanical Tech-

Installed in a Two-Fluid Two-Region MSBR — Effect on Liquid

Levels of Stopping One Pump During Normal Operatlon,
ORNL-MSR-67-108 (Dec. 22, 1967).

nology Incorporated, Apr. 12, 1968.

314, G. Grindell, Summary of Study of Feasibility of
Rotor-Bearing System for a 1250-hp Molten-Salt Fuel Pump
Conducted by MTI on Subcontract No, 2942, ORNL internal
correspondence MSR-68-97 (June 27, 1968).

 
ORNL-DWG &67-11817

46

 

 

 

Q g
= =
< 8 «
5 w
o0 1] u
£3 z &8 " Z
k= & om b Z
. w x = < . T -
o it v - = = N o O
o 2 o< o x =. Sz 2 wt
6 3 & o 5 = 9 gs # g
s 8 S50 @ o o o =8 £ T

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

L o}
WL Nz oH8

U1 48— WYL 1E2

 

 

WO U9

 

 

 

 

 

W8LY HIE—

 

 

 

UL Yib P

 

 

Fig. 5.8. Fuel Salt Circulating Pump,
 

bonded tungsten carbide, (2) nickel-bonded tungsten

nickel-chromium-bonded chromium carbide, and (4)
molybdenum-bonded tungsten carbide. Specimens of
these hard-surface coatings were obtamed but had not
been tested.

Because the fuel salt pump is a critical item in a
molten-salt reactor and the proposed design of pump
was considerably outside the range of our experience,

' some features of the pump were examined in detail.

The rotor dynamics were studied under a contract with
Mechanical Technology Incorporated.>®:>? Computer
studies were made of the conceptual designs to de-

47

- full-scale model of the pump rotor dynamic system to

carbide and mixed tungsten-chromium carbides, (3) |

evaluate the dynamic response experimentally. .
The studies®! indicated that approximately 0.025 in.
of bow in the middle of a uniformly bowed shaft and
approximately 0.019 in. of eccentricity between the
inner and outer diameters of a uniformly eccentric shaft

- could be accommodated at the shaft critical speed. The

termine flexural and torsional critical speeds and

flexural response to dynamic unbalance. The stability

characteristics of the bearing designs were also re-
viewed. The work, as summarized by Grindeil??
covered both 9- and 7.5-in.-OD shaft sizes. The first
critical speed for the larger shaft was about 700 rpm

" and for the smaller shaft about 560 rpm, both below

the design speed of 1200 rpm. The second shaft critical
speed was substantially above 1200 rpm for the larger
shaft and about 25% above it for the smaller. Serious
study of the problems of acceleration and deceleration
of both sizes of shafts through the first critical speed
was recommended. It was further recommended that

~ the pump shaft be designed to operate below the first

shaft critical speed to reduce the probability of low-
speed, high-amplitude whirl and the problems of tra-
versing the critical speed. This would require reduction
of the shaft length, lower design operating speeds, or
both. These objectives would be difficult to attain
without major revisions to the two-fluid MSBR deS1gn
described in this report.

A preliminary analysis of the undamped torsional

critical speed® indicated that the two torsional critical
‘speeds that might affect  pump operation could be

strongly dependent upon the electromagnetic torsional
stiffness of the drive motor. By changing some of the

- component dimensions, such as increasing the stiffness

values are limited by the bearing eccentricity. A bearing
eccentricity value of 0.95 was used. in making the
calculations. A survey was made of U.S. manufacturers
who could fabricate the 34-ft pump shafts to the
tolerances required.® One was found who expressed
confidence that shafts of the diameters and wall
thicknesses of interest could be produced with a
guaranteed -straightness from end to end of 0.005 in.
and with an OD-ID concentricity of 0.005 in. or better.
It was estimated that the cost would be relatively high,
however. A study of the effect of the tolerances on the
pump design and costs and on the dynamic balancmg
facilities required was in the planning stage.

As shown in Fig. 5.6, a startup tank is provided above
the fuel salt primary heat exchanger. The purpose of
the tank is to provide submergence for the pump as it is
started and the reactor is filled, the pumping capacity
of the pump being significantly greater than the transfer
rate of the salt from the drain tank. If the pump were
stopped, intentionally or otherwise, the fuel salt would
flow upward into the bottom of the startup tank and
then through the 5.in. overflow pipe to the fuel salt
drain tank. Since the tank is not provided with a
cooling system the fuel salt is not allowed to fill the
tank to more than a few inches in depth except during
startup, at which time the overflow pipe is closed.

Some cooling of the startup tank and the pump shaft

_is provided when the pump is operating. A small stream

of the outer pump casing and accounting for inherent

dampmg, it appeared possible that the pump could

- operate satisfacton]y between the first and second

critical speeds. If supercritical speed operation were
chosen, the study strongly recommended that a prac-
tical means for dynamic balancing of the shaft be
developed and that a rotor-dynamic evaluation simu-
lator be | constrilcted.' This 'sirnulatof WOuld, be a

 

 32p_W. Curwen, Rotor- Dynamtc Feasibility Study of Molten
Salt Pumps for MSBR Power Plants, MTI-67TR48 Mechamcal
Technology Incorporated, Aug. 6, 1967,

of fuel salt from the pump discharge passes partway up
through the. center of the pump shaft and then up

“through a narrow annulus between the shaft and a

cooling tube surrounding the shaft. Salt leaving the
annulus at the top spills back into the tank. Another
small stream of salt from the pump discharge flows into
the double wall of the tank bottom and upward in the
outer double wall to the top of the tank. The flow then

- turns downward through the annulus formed between

the center column of the tank and the pump casing.
Analysis of the shaft and tank wall heating was only
partly completed, particularly with regard to removal of

- afterheat. These and other aspects needed further
. study

The pumps are arranged so that the rotary element
can be replaced by remote maintenance techniques
without having to cut any of the salt piping. After the

 
 

motor housing has been set aside, the 2-ft-diam pump
casing can be withdrawn, carrying the upper bearing

and seal, the lower molten-salt-lubricated bearing, and
~ the impeller with it as a unit assembly. Although not

clearly evident in Figs. 5.6 and 5.8, the inlet salt pipe to

_ the pump is welded to the wall of the vessel surround-

ing the pump casing, and the 90° elbow, or-inlet flow
guide on the inside of the casing, is an integral part of
the rotary element and is withdrawn with it. Main-
tenance of the drive motor and upper bearing and seal
assembly can probably be performed in place through
use of a static seal on the shaft to isolate the upper
assembly from gasborne fission products and other
contammants in the pump tank and fuel salt system, -

5.4.3 Blanket Salt Circulating Pumps

There is little difference between the fuel salt and
blanket salt pumps except in the capacity and horse-

. power requirements, as shown in Table 5.4. The shafts

are about the same length, have the same bearing
arrangements, and the dynamic response is probably
similar.

5.4.4 Coolant Salt Circulating Pumps

The coolant salt pumps are similar to the fuel and
blanket salt pumps, although of larger capaclty, as
indicated in Table 5.4.

The coolant salt pumps are located near the top of .

the steam-generator cells and are of the short-shaft type

with an overhung impeller and do not need a salt-

lubricated lower bearing. These pumps will be operated

at variable speeds over the range from about 300 to - |

1200 rpm. Preliminary studies indicate that to operate
below the first critical speed the shaft would have to be

8 in. or more in diameter.

A double volute pump casing was selected for the
coolant salt pump in order to reduce the radial loads on
the impeller, particularly at off-design conditions. This
arrangement also reduces the diameter of the flexible
connection from the volute to the pump tank nozzle.

The coolant salt system would be provided with a tank

to act as a surge volume and to accommodate thermal
expansion of the coolant salt. :

5.5 Off-Gas System

Fission product gases must be continuously removed
from the circulating fuel salt to prevent '*$Xe from
absorbing so many neutrons that the breeding gain will
be significantly lowered. The neutron losses can be

greatly reduced by continuously spargingr the salt with

“helium which, in its subsequent removal, carries away
the xenon and krypton. Both of these gases are only

slightly soluble in the salt. Xenon that diffuses into the
pores of .the reactor core graphite must also be
considered. As discussed in Sect. 3.4, the amount of
xenon that diffuses depends on the ratio of the surface
area of helium bubbles in the circulating salt to the
surface area of graphite in the core, the rate of injection
and removal of bubbles, the coefficients for transfer of
xenon to both bubble and graphite surfaces, and the
permeability of the graphite to Xenon. Assuming a
processing-cycle time of about 1 min and 2 graphite

-coating effectiveness which reduces the permeability to
xenon as effectively as the preliminary tests indicate,

about 0.5 vol % of gas bubbles in the fuel salt in the
core will keep the !35Xe poison fraction below 0.5%.

The major features of the offgas system were

established, but only a few of the details were ex-
amined. The helium is injected into the circulating fuel

salt through a bubble generator at a rate of about 2.5 . -

scfm per reactor module. The bubble generator is a
Venturi-like section of pipe capable of producing bub-
bles with diameters in the range of 15 to 20 mils. The
bubbles recirculate with the fuel salt, making, on the
average, about ten passes through the primary system
before being removed by means of a centrifugal
separator. Swirl vanes at the entrance to the separator

induce rotational flow in the liquid that causes en-

trained gas bubbles to collect in an axial vortex from

which the gas is withdrawn. Vanes at the exit of the

separator remove the swirl.

The gas separator and bubble generator can be
installed in a bypass line around the fuel salt circulating
pump or in the main circulating system. In the former

location the bypass flow would have to be about 10%

of the total flow, the separator efficiency should be
near 100%, and a large pressure drop would be available
to operate the separator, If installed in the main stream,
the pressure drop that could be allotted to operation of
the separator would be much less, but the bubble

‘removal efficiency would only have to be about 10%.
. Effluent from the gas separator, composed of helium,

krypton, xenon, a “smoke” of noble metal fission
products, and as-much as 50 vol % of salt, is delivered.
to an entrainment separator. There it is joined by a

second stream of about 0.5 scfm of gas that is used to

purge the fuel salt pump tank. The salt and gas are
separated, the salt is returned to the suction of the fuel
salt circulating pump, and the gas with its burden of
fission products and a small amount of salt mist is
discharged to a gas processing system that retains the

e
»

xenon for about 48 hr — time enough for most of the

¥35Xe to decay — before it is recycled to the reactor.

Whether four, two, or one gas processing system
would be provided for a 1000-Mw(e) modular plant had
not been decided, but for the purposes of this discus-
sion we will assume that the gases from the four reactor
modules are combined and handled by one processing
system. The first part of the gas processing system is a
decay tank where the 12 scfm of gas is held for about 1

-hr and most of the short-lived fission products release

their heat. In this tank the solid daughters of the

49

radioactive gases, the noble metal particles, and the salt .

mist are separated from the gases by a particle trapping
system and are sent to the fuel reprocessing plant. Heat
is released in the decay tank and particle trap at a rate
of about 18 Mw, and care must be taken to provide
adequate cooling.

Gas leaving the 1-hr decay tank passes into beds of
charcoal that are designed to retain xenon for about 47
hr. These traps are water-cooled, and. the heat load is
about 3.2 Mw. On leaving the charcoal beds, about 10
scfm of gas passes through a water monitor and trap to
a compressor that recirculates the gas to the bubble

- generators in the reactor primary systems. The remain-

ing 2 scfm is processed further to remove the krypton,
Xenon, and tritium, and the helium is recycled to the

pump tanks and to other parts of the reactor that

require clean gas.

56 Dram Tanks

5.6.1 General

Drain tanks are provided for the fuel, blanket, and
coolant salts so that they may be safely stored and
isolated when maintenance or reactor replacement is

required. Draining of the fuel salt is also a shutdown

- measure in that the reactor quickly drains and becomes
 subcritical if the fuel salt pump stops. In any situation

where heat generated in the primary system could not
be effectively removed via the coolant salt circuit, it

would be necessary to quickly drain the fuel salt into’

the storage tank where an independent heat removal
system is provided.>*
The volumes of salt to be stored were not firmly

~ established because of dependence on only tentative

plant layouts, but a rough estimate of the total storage
requirements is 1200 to 1400 ft® of fuel salt, 2000 to

 

33The reference literaturé sometimes refers to the storage
tanks as “‘dump tanks” because of the quick-drain feature in the-

two-fluid MSBR.

2500 ft*> of blanket salt, and 800 to 1000 ft*> of
coolant salt. |
Two fuel salt tanks, four blanket salt tanks, and four

_ coolant salt tanks are provided for each reactor module.

These tanks are installed in a'drain tank cell that is
shared by adjacent reactor modules, as shown in Fig.
4.6. In addition, a flush salt tank (see Sect. 5.6.5)
located in the same cell serves both reactor modules.
The drain tank cells are heated to maintain the salts
above the liquidus temperatures.

The fuel salt drain tank represents more of a design
problem than the other salt storage vessels and will be
discussed in greater detail. It may be noted that the fuel
salt drain tanks have many of the design requirements

-of the reactor vessel itself in that the tanks must be

fabricated of Hastelloy N, are designed for essentially
the same pressures and temperatures, and must meet
the same requirements for leak-tightness and integrity.
In addition, each of the eight tanks must have sufficient
heat transfer surface for removing at least 12 Mw(t) of
heat from the drained fuel salt. Conceptual designs for
the drain tanks are presented here Many details
remained to be examined.

5.6.2 Fuel Salt Drain Tanks

The two fuel salt drain tanks are connected together
at the bottom of a salt line provided with a freeze valve,
as indicated on the flowsheet, Fig. 4.5. The pump
overflow line enters the top of one tank, and the system
drain line enters the bottom of the other. By using two
tanks, pressurization can be used to return the salt to
the circulating system without need for a valve in the
pump overflow line. The volume of the heels left in the
tank is also reduced.

Each of the fuel salt drain tanks is about 5 ft in
diameter X 25 ft high, as shown in Fig. 5.9. Pertinent

-data are given in Table 5.5. The salt-containing portion

is about 19 ft 6 in. high and has % -in.-thick Hastelloy N
walls. The 1-in.-thick inverted dished bottom head is

designed to minimize the inventory of fuel salt in the
tank heel. The inside of the tank has a Y4-in.-thick liner,

- or skirt, standing off from the wall about ¥ in., which

acts as a.downcomer on filling the tank and as a riser
when gas pressurization is used to empty it. It may be
noted in Fig. 5.9 that the riser skirt communicates with
the tank only at the bottom of the heel. A drain hne is
provided at the low point.

Steam at 500 to 600 psia and about 650°F is
introduced as a coolant at the top of the drain tank.
The steam enters through an 18-in.-diam reinforced
nozzle in the 1Y -in.-thick top head. The steam then

 
 

 

3'& in.

1% in.

el

&Y

Q072 in.

274 TUBES

STEAM (48-in. PIPE)
650 *F - 570 psig

ORNL-DWG 67 - 106414

 

 

 

 

 

 

 

 

 

 

 

0.042 in.

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2 in. 0D x Q035 in.
17 in, OD x 0.049m.
1% in. OD x 0.025 in.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
 

 

 

 

 

 

 

Fig. 5.9, Fuel Salt Drain Tank for Two-Fluid MSBR.

2% A PITCH y 1% .
. 44 DISH R TYP
ALL HEADS
€0 in. .
DlAMV rz in,
TS A Y S S SRS
' sfenu'tta-ln PIPE )
el .
. 44in
% UFFER'
' % §-w_ —§ g__ 6AS
3 Q’ \ § & N D g "%
% in. %—» L7 - -FUEL
7, . N— 11 K4 A4 | /5-in.PIPE)
- leain. NG L L Rl ( 24 £1-9%in.
N
[~ N o
N} N
‘ N\
N
\ \ PLENUM L
M <
N
: RISER SKIRT
\ . ’ E HEAT SHIELD
N _
3 LB e 19 ft-6in
\ I I 3 E SUPPORT
| \
N By
N _
N
N
N
3
N
3
FUEL DRAIN
1)

 

circulates through 271 cooling thxmbles which are
immersed in the fuel salt. The steam flows downward
through 1%-in.-OD X 0.025-in.-wall-thickness tubes

.which are inside a 1%-in.-OD X 0.049-in.-wall-thick-

ness tube to form an annular passage through which
the steam returns upward to the steam chest at the top
of the tank, ,

Each thimble is encased in a 2-in.-OD X 0.035-in.-
wall-thickness thimble which provides the requisite

51

double cdntainment between the fuel salt and the

~steam. The 0.027-in. annular space between the inner

and outer thimbles is filled with a stagnant salt,
probably of about the same composition as the coolant
salt, which acts as a heat transfer medium. While this
buffer space between the thimbles retards the heat

- transfer somewhat, it has the desirable effect of limiting

the thermal shock on the steam thimbles after a drain
and also of preventing excessive thermal gradients,

Table 5.5. Fuel Salt Drain Tank Data

 

Number required per reactor module
Rate of heat transfer per tank, Mw
Coolant in thimbles
Inlet coolant temperature, °F
Outlet coolant temperature, °F
Inlet steam pressure, psm
‘Fuel salt temperature, °F
- Number of thimble assemblies per tank
Active heat transfer length, ft
. Thimble spacing arrangement
Thimble pitch, in..
Shell inside diameter, in.
Steam flow rate, Ib/hr
Steam pressure drop, psi
Steam outlet velocity, fps
Thlcknesses, in.
Tank wall exposed to steam
Tank wall exposed to salt
Top dished head
Bottom reversed dished head
_Top tube sheet (flat)
Middle tube sheet (dished)
Bottom tube sheet (dished)

Thimble tubes, in.

" Quter wall
" Inner wall
Coolant supply tube

Calculated heat trainsfer coefficients,
- Btuhe ! °F 72
Fuel salt film
Outer wall
Stagnant salt
Inner wall .
Steam film
Overall

Calculated stresses at inside surface
of inner tube in bayonet assembly, psi
Hoop stress due to pressure
. Longitudinal stress due to pressure’
Radial stress due to pressure
Maximum thermal stress
Allowable stress intensity

Maximum pnmary plus secondary stress intens:ty

Approximate storage volume per tank, ft3

2

12
Steam
650
1000
540
1150
271
19%

A
2%,
60
211,000
7
72

ID : oD

1.930 2.000
1.777 1.875
1.450 1.500

130
4000
184
2180
189
52.3

10,400
-5200
-540
2760

.- 17,000

13,700

200

 

 
 

 

 

52

Table 5.6. Decay Heat of Fission Products in Fuel Salt

 

Time After

Watts per £t3 of Fuel

Difference in Heat

 

Drain Gross Amounts of Kr and Xe Sparged " Generation (%)
Fission Products on 30-sec Cycle ,
x 10° x 10%

0 , 16.4 14.4 _ 24.2

1 inin 6.2 47 14.2
Smin - ' 48 3.5 ‘ 27.1

10 min 4.2 3.0 28.6

30 min 3.1 2.2 . 29.0

1hr 2.5 1.8 - . 28,0

2hr ‘ 1.9 1.45 23.7

Shr : ‘ .35 1.08 20.0

1 day 0.699 0.656 6.1

 

The exit steam chest is formed by the uppermost
2-in.-thick tube sheet and an inverted dished head about

3Y, in. thick. A third 1% -in.-thick tube sheet forms the

buffer space for the stagnant salt. A ' -in.-thick heat
shield is suspended beneath the lower tube sheet to
protect it from thermal gradients and stresses when the
hot fuel salt enters the drain tank after a sudden drain.

Thimble support plates %4 in. thick are suspended from .

the lower tube sheet to minimize vibrations induced in
the thimbles by the flowing steam and to maintain the
spacing. ‘ »

- The two fuel salt tanks which serve a reactor module
are located in the deeper end of the drain tank cell, as

shown in Fig. 4.11, with one of the tanks at a higher

elevation than the other. The upper tank is the one
depicted in Fig. 5.9 and has a 5-in. salt drain line nozzle
at the top connected to the overflow from the fuel salt
circulating pump bowl. The drain tank at the lower
level does not require the 5-in. nozzle but is filled and
emptied through the 1-in. bottom drain connection.
This 1-in. line is connected through a freeze valve to the

“bottom of the fuel salt primary heat exchanger.

The liquidus temperature of the fuel salt is about
842°F, Although there is little danger of the fuel salt

The cooling steam in the drain tanks could be heated
to as high as about 1000°F in the thimbles by the
conditions existing immediately after a drain following
long-term operation at full reactor power. The steam
would be condensed in the turbine condensers, and the
condensate would be returned to the feedwater system. -
In this two-fluid MSBR concept other reactor modules
could continue to operate even though one or more of
the reactors had been drained. In the event that all the
reactors were drained, cooling steam would be supplied
by the auxiliary boiler which is used to supply initial
warmup steam for the plant. '

The heat generation in the fuel salt after a drain from
the reactor was investigated by Carter.>* He considered

both the equilibrium concentrations of fission products

freezing once the reactor has operated at power, -

nevertheless the cooling steam temperature cannot be
operated too far below the salt temperature if the
likelihood of local freezing of the salt is to be avoided.
Of greater concern are the temperature gradients in the
tube walls and tube sheets if the differences in
temperature ‘between the salt and the steam are too
great. The cooling steam has been assumed to be
admitted at about 650°F. The source of the 650°F

~steam has not been fully studied, but presumably it

could be taken from the exit of the reheat steam
preheaters in the turbine plant.

with no sparging of krypton and xenon during reactor
operation and the concentrations of fission products if
these gases were sparged from the reactor system on a
30-sec cycle. The results are shown in Table 5.6. '

"The heat transfer to be expected in the drain tank was
studied by Pickel.>® The results are summarized in
Table 5.5. Preliminary investigation of the stresses
indicated that they were within allowable limits. A
complete analysis of the vessel was not made, however.
Use of air rather than steam as a coolant was also
briefly investigated.

5.6.3 Blanket Salt Drain Tanks

A total of 16 drain tanks was selected to store the
estimated 2000 to 2500 ft® of blar!ket salt. This
provides four blanket salt tanks per module.

 

34w, L. Carter, Heat Generation in MSBR Fuel After

" Removal from the Reactor, ORNL-MSR-67-57 (July 31, 1967).

35T, W. Pickel, Heat Removal from (MSBR) Fuel Dump
Tanks, ORNL-MSR-67-72 (Sept. 6, 1967).
 

 

The amount of heat that could be generated in the
blanket salt after an emergency drain to the storage
tanks was not calculated, but the preliminary assump-
tion was that no cooling thimbles would be needed in
the tanks. If required, a steam cooling system similar to
that used in the fuel salt tanks would be provided.

5.6.4 Coolant Salt Dram Tanks

The leyouts of coolant salt piping were not suf-
ficiently detailed to estimate the quantity of coolant

‘salt in the systems. A rough estimate of the storage

capacity required was 800 to 1000 ft>, but this is likely
to be low. Four coolant salt tanks were provided per
reactor module. '

-The coolant salt tanks would not require cooling
systems.

5.6.5 Flush Salt Drain Tanks

A flush salt is provided for removal of residual fuel
salt from the circulating system in order to lower the
radioactivity level during maintenance and to assure
more complete recovery of valuable constituents. On

startup, the flush salt would be circulated in the

53

systems to sweep out foreign materials, moisture, etc., -

before introducing the enriched salts. The composition
of the flush salt would be very similar to the ?LiF-
BeF, fuel carrier salt. The volumes required and the
tank sizes were not established. '

5.7 Steam Generators

The 1000-Mw(e) MSBR power station described in this

report requires about 10 X 10° Ib/hr of total steam

generation. This is divided between 16 steam gen-
erators, or 4 steam generators per module. The number
of units was selected on the basis that the high (3800
psia) design pressure on the steamn side made larger

capacity units appear to have disproportionately thick -

heads and tube sheets. Maintenance aspects also favored
selection of a multiplicity of units since the generators

module. Load regulation and partial-load operation
received only superficial investigation.

- As shown in Fig. 5.10 the steam generator is a vertical
shell-and-tube unit with counterflow between the once-
through passage of the supercritical pressure water in
the tubes and the coolant salt in the baffled shell space.
The generator has a U-shaped cylindrical shell about 18
in. in diameter with each leg standing about 34 ft high,
including the spherical head. A baffle on the shell side
of each tube sheet provides a stagnant layer to help
reduce the stresses in the sheet due to temperature
gradients. The coolant salt can be drained from the
shell, but the water would have to be removed from the
tubes by evaporation, by gas pressurization, or by
flushing. (Drainability of the water was considered
desirable but not mandatory.) Both the tubes and shell
are fabricated of Hastelloy N in this design concept, but
less-expensive materials might be acceptable.

The principal data for the steam generators are listed
in Table 5.7.The design variables to be determined were .
the number of tubes, the tube pitch, length of tubes,
thickness of tube wall, thickness of tube sheet, baffle
size and spacing, diameter of shell, thickness of shell,
and thickness and shape of the heads. Because of the
marked changes in the physical properties of water as
its temperature is increased above the critical point at
supercritical pressures, the temperature driving force
and the heat transfer coefficient varied markedly along
the length of the tubes. These conditions required that
the heat transfer and pressure drop be calculated for
increments of length. An iterative procedure was
programmed for the CDC 1604 computer, as described
in ORNL-TM-1545.2% Based on the coolant salt prop-
erties given in Table 3.1, the optimum design was
calculated to have a long slim shell and relatively wide
baffle spacing as shown in Fig. 5.10. Subsequently the
thermal conductivity of the coolant salt was found to
be substantially less than had been used in the
calculations. Use of the lower thermal conductivity
could be expected to increase the number and length of

- tubes and to increase the shell d:ameter by a small.

as designed are not easily repaired and replacement of -

~ entire units could be required. \
The coolant salt flow is proportioned between the'

steam generators and the reheaters as necessary to
obtain a 1000°F outlet steam temperature from each.
About 87% of the total coolant salt flow is required for
the steam generators. The coolant salt is cooled from
about 1150 to 850°F in the units. Flow control is
accomplished either by a regulating valve in the salt
line, as indicated in Fig. 4.5, or by use of two
variable-speed coolant salt circulating pumps - per

’ amount.

5.8 Steam Reheaters
- A single-reheat power cycle was selected for the
MSBR plant although additional stages of reheat could
be provided should this prove to be economically
desirable. The steam conditions used in this study, and
shown in Fig. 4.12, are that the steam from the

high-pressure turbine exhaust is at 552°F, a tempera-
ture judged to be too low to be admitted directly into

 
 

 

   

SUPER CRITICAL
FLUID OUTLET

COOLANT SALT
INLET

 

 

54

-ORNL-DWG 65—12383A

     
 
  

SUPER CRITICAL
FLUID INLET

TUBE SHEET '

    
   

   
  

  

A-..........,.....h

il ‘iii’if;fj

I

i -

et COOLANT SALT
OUTLET :

  

TIE ROD AND SPACER

 

 

 

 

 

 

 
   

    

HHUWHINW‘“N _

:

BAFFLE

i
i

   

    

Fig. 5.10, Steam GenemtorSupahmmr.

the reheaters without the likelihood of local freezing of
the coolant salt. The steam is therefore preheated by
use of prime steam. (See Sect. 5.9 for a description of
the preheaters.) The preheated steam, at about 650°F
and 570 psia, is then reheated to 1000°F in the steam
reheaters. About 13% of the total reactor heat output is
used for steam reheating in a total of eight units, or two
per module. Selection of the number of units was
largely intuitive because optmnzatlon studles had not
commenced

The reheater units are counterflow, vertical, shell-
and-tube exchangers with straight tubes containing the
steam and coolant salt flowing through the disk and
doughnut baffles on the shell side. Tubes and shell are
constructed of Hastelloy N. The principal data are listed
in Table 5.8, and the unit is pictured in Fig. 5.11.

The methods used in the calculations of the heat
transfer and stresses are much the same as those used
for the steam generator and are descnbed in detail in
'ORNL-TM-1545.2% : -
 

9

 

Table 5.7. Steam Generator Data -

 

Type

Number required per reactor module

~ Rate of heat transfer, each

Mw
Btu/hr
Shell-side conditions
Hot fluid
Entrance temperature °F
Exit temperature, °F
Entrance pressure, psi
Exit pressure, psi
Pressure drop across exchanger, psi
Mass flow rate, Ib/hr
Tube-side condxtlons
Cold fluid
Entrance temperature, °F
Exit temperature, °F
Entrance pressure, psi
Exit pressure, psi
Pressure drop across exchanger, psi
Mass flow rate, lblhr
Mass velocity, 1b h: ft 2
Tube material
Tube OD, in.
Tube thickness, in.

Tube length, tube sheet to tube sheet ft

Shell material

Shell thickness, in.

Shell ID, in.

Tube sheet material

Tube sheet thickness, in.
Number of tubes

Pitch of tubes, in

Total heat ttansfer area, ft2
Basis for area calculation

“Type of baffle

Number of baffles
Baffle spacing
Overall heat transfer coefficient, U,
Btuhr™! ft™2 ,
Maximum stress intensity,? psi
Tube
Calculated
Allowable
- Shell ,
Calculated
Allowable
Maximum tube sheet stress, psi
Calculated
Allowable

U-tube U-shell exchanger with crossflow
baffles
4

120.9
4.13 X 108

Coolant salt
1125

850

252

194

58.1

3.6625 X 10°

Supercritical fluid
700 '
1000

3766.4

3600

- 166.4

6.3312 X 10°
2.78 X 10°
Hastelloy N
0.50

0.077

63.81
Hastelloy N
0.375

18.25

- Hastelloy N

4.75

349

0.875

2915

QOutside surface
Crossflow

9

Variabie

1030

P, =13,843;P, +(Q = 40,662 |
P, =S, =16,000;P, +Q =38, =48,000

P, =6372;P, +0=14420 .

Ppy =5y, = 10,500;P,,, + 0 = 35, = 31,500
<16,600

16,600

@The symbols are the same as those used in Sect, III of the ASME Boiler and Pressure Ves-

sel Code.

 
 

 

 

Table 5.8, Steam Reheater Data

 

Type

Number required per reactor module
Rate of heat transfer per unit
Mw ' .
Btu/hr
Shell-side conditions
Hot fluid
Entrance temperature, °F
Exit temperature, °F
Entrance pressure, psi
Exit pressure, psi
Pressure drop across exchanger, psi
Mass flow rate, Ibfhr
Mass velocity, Ib hr ! £t~

Tube side conditions

Cold fluid
Entrance temperature, °F
Exit temperature, °F
Entrance pressure, psi
Exit pressure, psi
Pressure drop across exchanger, psi
Mass flow rate, Ib/hr
Mass velocity, Ib/hr ™! £t 72
Velocity, fps
Tube material
Tube OD, in.
Tube thickness, in.
Tube length, tube sheet to tube sheet, ft
Shell material
Shell thickness, in.
Shell ID, in.
Tube sheet material
Tube sheet thickness, in.
Number of tubes
Pitch of tubes, in.
Total heat transfer area, ft?
Basis for area calculation
Type of baffle o
Number of baffles
Baffle spacing, in.
Disk OD, in.
Doughnut ID, in,
Overall heat transfer coefficient, U
Btuht™! £t72
Maximum stress intensity 4 psi
Tube
Calculated
Allowable
Shell
Calculated
Allowable
Maximum tube sheet stress, psi
Calculated .
Allowable

Straight tube and shell with disk and dough-
nut baffles
2

36.25
1.24 x 10®

Coolant salt
1125

850

208.5

197.1

114

1.1 X 10°
1.44 X 10°

Steam

650

1000

580

568

12

6.3 x 10°
3.98 X 10°
145
Hastelloy N
0.75

0.035

22.1
Hastelloy N
0.5

28
Hastelloy N
4.75

628

1.0

2723

.Outside of tubes

Disk and doughnut
10 and 10 '
12.375

24.3

16.9

285

P, =4349;P, +(Q = 13,701 |
P, =S, = 14,500;P,, +Q = 3S,, =43,500
P,, = 6046.5; P, + 0 = 17,165

Py = Sp, = 10,600;P,,, + 0 = 35, = 31,800

<10,500
10,500

@The symbols are the same as those used in Sect. Il of the ASME Boiler and Pfessur_e Ves

sel Code.
ORNL—DWG 65—12381A

STEAM OUTLETi |

 

 

 

 

 

 

 

 

 

   
  
   
  

./éé"
O TUBE SHEET
> r'f‘-':‘.’-':" )
hl
i
i
SALT i TIE ROD
oI
‘ § \ SPACER
LW . et
il )
'lf; 1 .
o b
|5f i ilh
H )
' ;! P
I Al
TUBULAR g (i L)
’E! P
i il
i Vet
I!E‘ i L i
TUBES
3
DISK BAFFLE
~ DOUGHNUT
BAFFLE
msuufignow ' "SALT
BAFFLE OUTLET

_
%

\\DRA;N LINE

|} sTEAM INLET

*

Fig. 5.11. Steam Reheater.

57

 

5.9 Reheat Steam Preheaters

Steam at turbine throttle conditions of 3500 psia and
1000°F is used to preheat the reheat steam from 552 to
650°F before it enters the reheaters. The eight pre- -
heaters, two per module, are single-pass, counterflow,
U-tube, U-shell units with the supercritical-pressure
steam in the tubes and the reheat steam in the
unbaffled shell, as shown in Fig. 5.12. Selection of a U
shell rather than a divided cylindrical shell permits
smaller diameters for the heads and reduces the
thickness required for the heads and tube sheets.

- Principal data are given in Table 5.9. The heat transfer

and stress calculations are covered in ORNL-
TM-1545.28 '

The preheaters are more a part of the turbine plant
than the reactor plant and need not be installed in a

shielded cell nor necessarily manifolded in conformity

with the modular arrangement adopted for the reactor
plant. Two preheaters have been shown associated with
each reactor module, however, primarily as a matter of
convenience in the layout.

5.10 Maintenance

Maintenance is a major subject for consideration in
the design of any fluid fuel reactor, and it is discussed
briefly here only because the two-fluid MSBR con-
ceptual studies were discontinued before maintenance
procedures could be considered in detail. It was,
however, recognized throughout the design effort that
it must be possible to repair or replace system com-
ponents within a reasonable downtime for the plant,
and this requirement influenced much of the design.
Even though the systems containing fuel salt are
drained and flushed, the residual radioactivity will
require that all maintenance be accomplished by re-
motely operated tools and equipment. The off-gas
systems will also require remote maintenance. On the

other hand, most of the coolant salt system com-

ponents can probably be approached for direct main--

~ tenance after flushing and elapse of a short decay time.

Little or no radioactivity should be present in the steam
and feedwater systems even during full-power opera-
tion. , - ,

As mentioned in Sect. 2, the radiation damage to
graphite will make it necessary to replace the reactor
core vessel several times during the lifetime of the plant.
Since the two-fluid MSBR concept does not lend itself
to use of a removable top cover for the reactor vessel to

- gain access for replacement of the core graphite, it was

decided that use of four small r'_eac'tors or modules, with
replacement of an entire reactor vessel and core

 
 

 

assembly, would be more practical than in-place main-
tenance of a single, larger reactor. Replacement of a
- module would require cutting of the salt piping and
withdrawal of the assembly upward into a shielded

transport cask for transfer of the spent unit into a -

shielded pit for decay and ultimate disposal. A shop-
assembled and -tested replacement module would be

- standing by. The salt-piping stubs would be previously

machined for welding through use of a jig which
matches the installed piping system.

Table 5.9. Reheat Steam Preheater Data

 

Type

Number required per reactor module
Rate of heat transfer, each
Mw
Btu/hr
- Shell-side conditions
Cold fluid :
Entrance tempcrature, °F
Exit temperature, °F
- Entrance pressure, psi
Exit pressure, psi
Pressure drop across exchanger, psi
Mass flow rate, Ib/hr
Mass velocity, Ib hr ™! ft ™
Tube-side conditions
" Hot fluid
Entrance temperature, °F
Exit temperature, °F
Entrance pressure, psi
Exit pressure, psi
Pressure drop across exchanger, pst
Mass flow rate, lblhr
Mass velocity, Ib he™? ft
Velocity, fps
Tube material
Tube OD, in.
Tube thickness, in.
Tube length, tube sheet to tube sheet, ft
Shell material
Shell thickness, in.
Shell ID, in.
Tube sheet material
Tube sheet thickness, in
Number of tubes
Pitch of tubes, in
Total heat ttansfer area, £t
Basis for area calculation
Type of baffle
~ Overall heat transfer coefficient, U
Btuhr™? £t
Maximum stress mtensxty, psi
Tube
Calculated
Allowable

Shell
Calculated
Allowable

‘Maximum tube sheet stress, psi
Calculated
Allowable

One-tube-pass, one-shell-pass U-tube, U-
shell exchanger with no baffles
2 - . .

12.33
4.21 X 107

Steam

551

650

595.4

590.0

54

6.31 X 10°
. 356 x 10°

Supercritical water
1000
869
3600
3535

- 65
3.68 X 10s
1.87 X 10°®
93.5
Croloy
0.375
0.065
13.2
?rqloy

16

20.25
Croloy
6.5
603
0.75
781
Tube OD
None
162

P,,=10,503;P,, +0 = 7080

P, =S, "lO,SOOat%lFP +Q0=
38, ~31,500

-14375P +Q = 33,081
P 1sooomssopp +Q=
3s -45000
7800
7800 at 1000°F

 
 

qx

       

' SUPER CRITICAL
FLUID OUTLET

—

STEAM INLET
fi

"
e

 

 

ORNL OWG 65-123682A

SUPER CRITICAL
FLUID INLET

STEAM
OUTLET
#

|

Fig. 5.12. Reheat Steam Preheater.

If 2 major tube leak should occur in the primary heat
exchanger, it would be necessary to replace the entire
heat-exchanger—pump assembly. The procedure would
be to cut the large fuel salt pipes and the two inlet
coolant salt pipes, to cut the seal welds, and unbolt the
large flange at the bottom of the shell. The exchanger
could then be lifted from the cell, disengaging the
central coolant salt pipe at the slip joint provided for

~ this purpose. Drain, fill and drain, gas pressurization, -
‘and several other connections must aIso be cut when

removing the exchanger.

The rotating parts of the fuel salt clrculatmg pump
.can be withdrawn upward -after the drive motor has

been set aside. This is a relatively simple operation that

- does not require breaking the salt piping connections.

The type of maintenance of a large MSBR reactor

plant described here requires the cutting and welding of -

salt piping by remote means, Some original work by the
Air Force has been modified and is being developed by

Holz3® at ORNL to provide this capability. A compact

orbital system is designed to be clamped around the
pipe and has interchangeable modules and a weld

programmer for cutting, beveling, tungsten-arc welding, -

and inspection. Preliminary tests have produced welds
of acceptable quality in 6- and 8-in.-diam pipes with
fully remote operation.

Much valuable experience has been gained at the
MSRE with remote maintenance. ‘operations similar to
those required for a larger molten-salt reactor. The use
of jigs and optical tooling has proven to be a practical
and expeditious method of fitting replacement parts
and components into the existing system. :

The first.cost of the special equipment required for
maintenance operations is a part of the capital cost of

‘the plant. This has been included in Table 7.1 as an
- allowance, since conceptual designs for the equipment

were not available. The cost of the materials used in
replacement of reactor modules and the special labor

. forces required are included in the power production
~cost as a separate item (see Table 7.2). (Some may wish

to include this expense with the fuel-cycle cost; others
may consider it to be an operating cost.)

 

36peter P. Holz, Feasibility Study of Remote Cutting and
Welding for Nuclear Plant Maintenance, ORNL-TM-2712 (No-
vember 1969).

 
 

 

 

6. REACTOR PHYSICS AND FUEL
CYCLE ANALYSIS

6.1 Optimization of Reactor Parameters

- In addition to the so-called conservation coefficlent

discussed in Sect. 2, which relates specific inventory

and breeding gain, two other principal indices by which -

the performance of a molten-salt breeder reactor can be
evaluated are the cost of power and the annual fuel
~ yield, The latter two indices were used as figures of
_merit in assessing the influence of various design
~ parameters and the effect of design changes on the
two-fluid MSBR.

We customarily combine the cost factor and the fuel -

yield, that is, the annual fractional increase in the
~inventory of fissionable material, into a composite
 figure of merit

F=y+100(C+X)™",

. in which y is the annual fuel yield in percent per year, C
is that part of the power cost which depends on any of
the parameters considered, and X is an adjustable
- constant, having no physical significance, whose value
merely determines the relative sensitivity of F toy and
C. Since a large number of reactor parameters are
involved, we make use of an automatic search pro-
cedure, carried out on a computer, which finds that
combination of the variable design parameters that
maximizes the figure of merit F subject to whatever
constraints may be imposed by the fixed values of other
design parameters. This procedure, called
OPTIMERC,®” incorporates a multigroup- diffusion

calculation (synthesizing a two-space-dimensional de-

scription of the flux by alternating one-dimensional
flux calculations), a determination of the fissile, fertile,
and fission product concentrations consistent with the
processing rates of the fuel and fertile salt streams, and
a method of steepest gradients for optimizing the values
of the variables. By choosing different values for the
" constant X in the figure of merit F, we can generate a

curve showing the minimum cost associated with any

~ attainable value of the fuel yield. By carrying out the
optimization procedure for different successive fixed
values of selected design parameters, we obtain famxhes
of curves of C as a function of y.

One of the design parameters which has a mgmficant

influence on both yield and power cost is the power

 

37In OPTIMERC any of some 20 parameters may be either -

assigned fixed values or be allowed to vary within speclf‘ ied
limits subject to the optimization procedure.

FUEL CYCLE CQST I_ mi!ls/kwhr(elecmcnl)]

60

 

ORNL-DWG 6711806

0.75 10 w/t:m3

 

 

 

 

0.50

 

0.25

 

 

 

 

 

o 2.5 50 _ 75
FUEL YIELD (% per year)

Fig. 6.1. MSBR Fuel Cyde Cost vs Annual Fuel Yield.

density in the core. The performance of the reactor is
better at high power densities. At the same time, the

useful life of the graphite moderator, which is de-

pendent on the total exposure to fast neutrons, is

inversely proportional to the power density (see Table

5.1 and Sect. 6.2). It is necessary, therefore, to

determine the effect of power density on performance -

with considerable care.
In Fig. 6.1 the fuelcycle cost is used because it
reflects most of the variations of power cost due to the

influences of the parameters being varied. It may be
seen from Fig. 6.1 that a reduction in average power

density from 80 to 20 w/cm® involves a fuel-cycle cost
penalty of about 0.1 mill/lkwhr(e) and a reduction in
annual fuel yield of perhaps 1.5%. There is an increase
in the capital cost of the reactor, but this is offset

somewhat by a reduction in the cost of replacing the-
~ graphite (and the reactor vessel) since this can be done

at less frequent intervals. The penalty for having to
replace the graphite (compared with a high-power-

density core not requiring replacement) is about 0.2

mill/kwhr(e). The capital cost portion increases and the
replacement cost portion decreases with decreasing
power density so that the total remains about constant.

o
 

»

»

61

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.20 l
- | saLT voLumE FRacTIONS
3 ) e ey —
2 T L
& - . FISSILE
. i’ . .
S o2 6/ _
J #
3 10 / /
S
.,
5 .
& oo8 '
10_-;-—-1':— = FEH;‘E\_—
g/ ‘ TP == EoTTa
2 — —
—, 004
g ,
5 |
.g /¢_
= o / _ FUEL SPECIFIC POWER
w / ‘
E L | |
Q
w
L
w
o
“w 0
o 20 40 60 80

POWER DENSITY (w/cm3)

ORNL-DWG 67-{1BO7R

 

0 -
: \ CARBON-TO-URANIUM RATIO

 

S

 

 

¢/u RATIO (x107)
S
//
/ P
/ v
K/

 

 

 

 

 

 

 

6— v—-___ Ty
\-._____:
4 .
0 20 40 60 80

POWER DENSITY (w/cm®)

Fig. 6.2. Variation of MSBR Parameters with Average Core Power Density, Numbers attached to curves are values of the

adjustable constant X.

Figures 6.2 and 6.3 show the variation of other selected

parameters with power density and the adjustable

constant X. For given values of power density and X,

the corresponding values of the selected parameters are
those of the reactor with the optimum combination of
yield and fuel-cycle cost.

It is apparent from these results that the useful life of
the graphite is not increased by reducing core power
density without some sacrifice in other aspects of

“reactor performance. The reduction in yield and the
increase in cost are quite modest for a reduction of
power density from 80 to 40 w/cm?, but they become

increasingly more significant for each further factor of
2 reduction in power density. Nonetheless, as shown in

~ Fig. 6.1, it appears that with an average power density

as low as 20 w/cm® the MSBR can still achieve an
annual fuel yield of 3.5 to 4% and a fuel-cycle cost of
about 0.5 mill/kwhr(e). \

The fuel-cycle cost estimate for the 40-wlcrn .con-

figuration summarized in Fig. 6.1 is shown in more

detail in Tables 6.1, 6.2, 6.3, and 6.4. The economic
ground rules for the fuel-cycle cost calculations are
given in Table 6.1. The worth of the fissile isotopes was

‘taken from the AEC price schedules.. The capital
changes of 13.7%/year for depreciating items and
10%/year for nondepreciating materials are typical. of
‘those for privately owned plants under 1968 condi-
tions, as shown in Appendix Table A.12.

Results of the fuel-cycle calculations for the MSBR
design ‘are summarized in Table 6.2, and the neutron
balance is given in Table 6.3. The reactor has the

Table 6.1. Basxc Economic Assmnpuons Used in Nuclear Design

 

. Studies
Reactor power, Mw(e) , o 250 . .
Thermal efficiency, % - 45
Load factor : ‘ ; - 0.80
Cost assumgtxons .
Value of 3U and 233Pa $/lg 14.00
Value of 235U, $/g . 12,19
Value of thorium, SIkg 12.37
Value of carrier salt, $/kg 25.97
Capital charge, %Iyear '
Plant . _ 13.7

Nondepreciating capital, including fissile inventory ~ 10.0

 

 
 

THICKNESS (ft)

0.08

0.06

0.08

0.06

NEUTRONS ABSORBED/7 SOURCE NEUTRONS

0.02
2.25

2.24

'7(

2.23

2.22

400

43

200

'Fig. 6.3. Variation of MSBR Parameters with Average Core
Power Density, Numbers attached to curves are vatues of the

ORNL-DWG 67~-11808R

- RADIAL BLANKET THICKNESS

 

 

NEUTRON LOSSES TO Li,Be,F -

_— e e ol

— o -

 

 

 

 

 

 

 

 

 

—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

POWER DENSITY (w/cm3)

adjustable constant X,

ne
—
\
.
V'
whfi -
\ \-.,__.. 10
\ 6
- 2
0
6,10 FUEL SALT VOLUME
| ‘2
0NN ]
Tom\\ \

\
6,10 T
| \\‘ %f’—___h

0 Q\ _ ]
—

Cm::_-— —_—

|’ ,

20 40 60 80

- 62

~ Table 6.2. MSBR Fuel-Cycle Petformanee

 

 

 

 

 

 

 

Fuel yield, %/year 407
Breeding ratio 1.06
Fissile losses in processing, atoms per fissile absorptxon 0.0040
Neutron production per fissile absorption (7€) 2.22
Specific inventory, kg of fissile material per megawatt
of electricity produced 1.26
Specific power, Mw(t) per kg of fissile material 1.77
Power density, core average, kw/liter
Gross 19
In fuel salt 140
Fraction of fissions in fuel stream 0.996
Fraction of fissions in thermal-neutron group 0.846
Mean n of"""’u o 2.225
Mean n of 235U 1.981
Table 6.3. MSBR Neutron Balance for Average -
' Power Density of 20 w/cm> ,
Neutrons per Absorption
: in Fissile Fuel
Matt.enal Total Alésitli":ed Neutrons
Absorbed VIS produced
. Fission
232 0.9876 0.0020 0.0047
233p, 0.0002
233y 0.9290  0.8267 2.0670
234 0.0801  0.0003 0.0010
235 0.0748 0.0607 0.1482
236y 0.0082 0.0001 0.0001
2375p 0.0011
238y 0.0
Carner salt (except 61i) ~ 0.0682 0.0210
61i : - 0.0068
Graphite 0.0430
35Xe. 0.0050
1496m 0.0061
151gm 0.0019
Other fission products 0.0187
Delayed neutrons lost? 0.0033
' Leakage 0.0080 ‘
Total 2.2420 0.89 - 224

 

"Delayed neutxons emitted outside the core.
Leakage mcludmg neutrons absorbed in the reflector.

advantage of no neutron losses to structural materials in
the core other than the moderator. Except for some
unavoidable loss of delayed neutrons in the external
fuel circuit, there is almost zero neutron leakage from
the reactor because of the thick blanket. The neutron

integrated processing.

losses to fission products are mmnmzed by the rapid

The portion of the fuel-cycle cost due to processing
losses is shown in Table 6.4 and is based on a fertile
 

»

63

Table 6.4, Estimated Fuel-Cycle Cost for a
Privately Owned Two-Fluid 1000-Mw(e) MSBR Power Station

 

Cost [mills/kwhr(e)]

 

Fuel Fertile Grand
Stream Stream Subtotal - Total

 

Fissile inventory?
- Thorium inventory
~ Salt mventory
Total mventory
Thorium replacement
Salt replacement
Total replacement '
Fixed charges for processing eqmpment"
Operating labor and supplies?
- Total o
Production credit®

Net fuelcycle cost, miflsli&whr(e) _

0.2242 0.0215 0.2457
0.0379 0.0379
0.0289 0.0482 0.0771
Lo 0.361
10.0043 0.0043
0.0491 0.0038 0.0529
: 0.057
0.10
0.11
0.628
—0.084

0.5-0.6

 

S[ncluding 233Pa, 233U, and 235U,

VbBased on 80% plant factor and fixed charges of 10%/year.

©Based on total equipment cost of $5.3 million taken from 0RN1.-39961 plus 10% esca-
lation of 1966—68 less $556,000 for structures included in Table 7.1, and fixed charges of »

13 7%Iyear with 80% plant factor.

dBased on operating costs of $793 ,000/year taken from 0RNL-39961 plus 10% escala-

* tion 1966—68 and 80% plant factor.

" €Based on 4% yield at average core power density of 20 w/cc.

'matenal loss of 0. 1% per pass through fuel-recycle

processing.

The fuel-cycle costs for fixed charges on processing
equipment are based on cost estimates published in
ORNL-3996, but escalated by 10% to 1968 conditions.
The operating costs for labor and plant supplies (other
than ‘salt inventories and makeup) specifically related to
the chemical processing portion of the power station

‘are also based on the ORNL-3996 estnnate with 10%

escalation, as shown in Table 6.4.

It may be noted in Table 6.4 that the main cost items
are for the fissile inventory and the processing costs.
The inventory costs are rather rigid for a given reactor
de51gn, since- they are largely determined by the fuel
volume externa] to the reactor core reglon The

Table 6.5. Processmg (,Yc!e Tlmes w:th

 

 

 

| X p
Powéi De.nslity;, _ CYCIC Time (days)
: _(W/Cm?) ~ FuelStream  Fertile Stream = Pa
80 - 50 , - 50 0.5
40 o 0 07
20 110 110 . . 11
10 173 144 14

 

processing “costs are, of course, a function of the
processing-cycle times, one of the chief parameters
optimized in this study. The processing cycle times for
the optimized case with X = 2 are given in Table 6.5.
The cycle times show a systematic increase with
decreasing power density.

- 6.2 'Useful Life of Moderator Graphite -

Information used in the two-fluid MSBR studies
regarding the - dependence of graphite dimensional
changes on. fast neutron dose was derived primarily
from expenments carried out in the Dounreay Fast
Reactor (DFR).

“A curve of volume change vs fast neutron dose for a
_nearly 1sotrop1c graphlte at temperatures in the range
/550 to600°C is shown in Fig. 6.4, which is taken from
the paper of Henson, Perks, and Simmons.?® The

neutron dose in Fig. 6.4 is expressed in terms of an

 

38R. W. Henson, A. J. Perks, and J. H. W, Simmons, Lattice -
-Parameter and Dimensional Changes in Graphite Irradiated

" Between 300 and 1350°C, AERE-R5489, to be published in the

proceedings of the Eighth Carbon Conference.’

 
 

 

14

 

ORNL-DWG 67-44809

 

 

2
¢ '400-440°C

550-600°C

[

 

10
o

 

o

 

H

 

n

 

o

 

- VOLUME CHANGE (%)

 

Q®

&
o .
o\\ o
_ - *
\& .

o o 0
0 ' K

EQUIVALENT PLUTO DOSE (41022neuirons/cm?)

 

 

 

 

 

 

 

 

Fig. 6.4. Volume Changes in Near-Isotropic Graphite Result-
ing from Neutron Irradiation. See text for dose in forms of
MSER flux.

equivalent Pluto dose; the to.tal DFR dose, that is,
P
S [ e ndEa,

is 2.16 times the equivalent Pluto dose. From an
inspection of all the available data, we concluded that a
dose of about 2.5 X 10?? neutronsfcm?® (equivalent
- Pluto dose) could be sustained without any significant
deterioration of the physical properties of the graphite.
This was adopted as the allowable dose in these MSBR
~ studies,
- mechanical design problems that might be associated
with dimensional changes in the graphite.

~ In order to interpret these experiments to obtain
predictions of graphite damage vs time in the molten-
salt reactor, it is necessary to take into account the
difference between the neutron spectrum in the DFR
and in the MSBR. This, in turn, requires assumptions

64

regarding the effectiveness of neutrons of different
energies for producing the observable effects with
which one is concerned. At present the best approach
available is to base the estimates of neutron damage
effectiveness on the theoretical calculations of graphite
lattice displacements vs-carbon recoil energy carried out
by Thompson and Wright.?® Their “damage function”
is integrated over the.distribution of carbon recoil
energies resulting from the scattering of a neutron of a

~ given energy, and the result is then multiplied by the

energy-dependent scattering cross section and inte-
grated over the neutron spectrum in the reactor. Tests
of the model were made by Thompson and Wright by
calculating the rate of electrical resistivity change in
graphite  relative to the 53Ni(n,p)*®Co reaction, in
different reactor spectra, and the data were compared
with experimental determinations of the same quanti-
ties. The results indicate that the model is at least useful

~ for predicting relative damage rates in different spectra.

pending further detailed consideration of

The spectral effects are discussed more fully by Perry in
ORNL-TM-2136.!2 :

A useful simplification arises from the observation
that the damage per unit time is closely proportional to
the total neutron flux above some energy Ey ,where E
has the same value for widely different reactor spectra.
We have reconfirmed this observation to our own -
satisfaction by comparing the calculated damage per
unit flux above energy E, as a function of E, for

- spectra appropriate to three different moderators (H, O

D, 0, and C) and for a “typical” fast reactor spectrum.
The results plotted in Fig. 6.5 show that the flux above
about 50 kev is a reliable indication of the relative
damage rate in graphite for quite different spectra.
Figure 6.6 shows the spectra for which these results
were derived. The equivalence between MSBR and DFR
experiments is found by equating the doses due to
neutrons above 50 kev in the two reactors. We have not’
yet calculated the DFR spectrum explicitly, but we
expect it to be similar to the “fast reactor” spectrum of
Fig. 6.6, in which 94% of the total flux lies above 50

‘kev. Since the damage flux in the MSBR is essentially

proportional to the local power density, we postulate
that the useful life of the graphite is governed by the
maximum power density rather than by the average,
and thus depends on the degree of power flattening that
can be achieved (see Sect. 6.3). In the two-fluid MSBR
the average flux above 50 kev is about 0.94 X 10'%

neutrons cm > sec™' at a power density of 20 w/em®.

 

39\, W. Thompson and S. B. Wright, J. Nucl. Mater. 16,
14654 (1965). . - ,
65

ORNL DWG &7-14810

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

"
/4
/ .
2 n
A
10 /7 3
I PH(E)DIE)IE 1/ 1
= .9 /
R(Eo) = —9 / /‘
. I "$(E)dE . . A
£, /
- 9 L ] f y /
% / F4 {
- /
= / /
5 1 = H,0 SPECTRUM )
L 8 2 /) / /
€ 2 = D,0 SPECTRUM ] 7
8 3 = CARBON SPECTRUM - ) /
2 4 = FAST SPECTRUM /7 /
o ‘ 7 /
° 7 ~23% ya. L
Ng /// Py
= 7 -
’B 44/ 7 -
) A —
x 6 3o, = -'{./
-2
| [ >
. ?1 ©
T — ] =
e - e
5 - .--—"-_—-- /’ O_
ot = /a/
- ®
’-’ _.‘
__,——’ o
4 |
10? 5 6
2 5 10 2 5 40
&s (eV)

Fig. 6.5. Fast Flux as a Measure of Radiation Damage.

In the DFR irradiations the equivalent Pluto dose of 2.5
X '10*? neutrons/cm? that was taken as the tolerable
exposure for the graphite is a dose of 5.1 X 10?2
neutrons/cm?® (50 kev).*® The approximate useful
lifetime of the graphite is then easily computed and is
shown in Table 6.6 for various combinations of the
average power density and peak-to-average power
density ratio.

It must be acknowledged that some uncertainties
remzain in applying the results of DFR experiments to

 

401 subsequent studies of one-fluid reactors the desxgn

 

 

Table 6.6. Useful Life of
MSBR Graphite
Average Power ' Life
- Density . Py, /Pg, -(full-power

(wjem?) “years)
40 2.0 43
40 1.5 5.7

- 20 20 ‘ 8.6 .
20 , 1.5 . 11.5

 

the MSBR, including the possibility of an appreciable

- dependence of the damage on the rate at which the

lifetime was lLimited to a fluence of 3 X 1022 neutronslcm’ E

> 50 kev) on the basis that expansion -of the graphite much
beyond the initial volume might increase the permeability to
salt and to account for the more rapld changes that occur at the
higher temperatures of 700 to 720°C in thé graphite. More
recent data (July 1969) seem to confirm that the lower fluence
is a better value for graphite obtainable in the near future.

dose is accumulated, as well as on the total dose. The
dose rate in the DFR was approximately ten times
greater than that expected in the MSBR, and if there is
a significant dose-rate effect, the life of the graphite in
an MSBR might be appreciably longer than shown in
Table 6.6. '

 
 

 

 

10 8- 6 4 3 2 1.5

ORNL-DWG 67-ti811

NEUTRON ENERGY (MeV) :
10 08 0.6 04 03 0.2 0.5

 

.5 040 0.08
; | 1 i | { ] { 1 I | I - | | ]
1=H,0 SPECTRUM
2=D,0 SPECTRUM
3=CARBON SPECTRUM . | .
4=FAST SPECTRUM (50% Na, 50% U-METAL, 20% 2>°u, 80% 2*%u)
1.0

 

¢ (V) (arbitrary units)

 

 

 

 

 

 

 

 

 

 

 

LETHARGY U

Fig. 6.6. Neutron Flux per Unit Lethargy vs Lethargy. Normalized for equal damage in graphite. -

6.3 Flux Flattening

Because the useful life of the graphite moderator in
the MSBR depends on the maximum value of the
damage flux rather than on its average value in the core,
there is obviously an incentive to reduce the maximum-
to-average flux ratio as much as possible, provided that
this can be accomplished without serious penalty to
other aspects of the reactor performance. In addition,
there is an incentive to make the temperature rise in
parallel fuel passages through the core as nearly uniform
as possible, or at least to minimize the maximum
deviation of fuel outlet temperature from the average

" value. Since the damage flux (in effect, the total
neutron flux above 50 kev) is essentially proportional
to the fission density per unit of core volume, the first
incentive requires an attempt to flatten the power
“density per unit core volume throughout the core, that
is, in both radial and axial directions. Since the fuel
moves through the core only in the axial direction, the
second incentive requires an attempt to flatten, in the

radial direction, the power density per unit volume of
fuel. Both objectives can be accomplished by main-
taining a uniform volume fraction of fuel salt through-

~out the core and by flattening the power density

distribution ‘in both directions to the greatest extent
possible. ' o |
The general approach taken to flattening the power
distribution is the classical one of providing a central
core zone with k_ = 1, that is, one which is neither a -
net producer nof a net absorber of neutrons, sur-
rounded by a “buckled” zone whose surplus neutron
production just compensates for the neutron leakage
through the core boundary. Since the fuel salt volume
fraction is to be kept uniform throughout the core and
since the concentrations of both the fuel and the fertile
salt streams are uniform throughout their respective
circuits, the principal remaining parameter that can be
varied with position in the core to achieve the desired
effects is the fertile salt volume fraction. The problem
then reduces to finding the value of the fertile salt
volume fraction that gives k., = 1 for the central,
 

 

 

 

flattened zoné, with fixed values of the 6ther param-.

eters, and finding the volume fraction of the fertile salt
in the buckied zone that makes the reactor critical for
different sizes of the flattened zone. As the fraction of
the core volume occupied by the flattened zone is

increased, the fertile salt fraction in the buckled zone

must be decreased, and the peak-to-average power
density ratio decreases toward unity. The largest flat-
tened zone and the smallest power density ratio are
achieved when the fertile material is removed entirely
from the outer core zone. Increasing the fuel salt
concentration or its volume fraction (with an appropri-
ate adjustment of the fertile salt volume fraction in the
flattened zone) would permit a still larger flattened
zone and smaller P, . /P, ., but this could be expected
to compromise the reactor performance by increasing
the fuel inventory.

There are many possible combinations of parameters
to consider. For example, it is not obvious, a priori,

whether the flattened zone should have the same

height-to-diameter ratio as the entire core, or whether
the axial buckled zones should have the same composi-
tion as the radial buckled zone. While we have by no
means completed investigations in this area, we have
progressed far enough to recognize several important
aspects. , o

First, by flattening the power to various degrees in
the radial direction only and performing fuel-cycle and
economic calculations for each of these cases, we find
that the radial power distribution can be markedly
flattened with very little effect on fuel cost or on
annual fuel yield. That is, the radial peak-to-average

‘power density ratio, which is about 2.0 for the uniform
~core (which is surrounded by a heavily absorbing
blanket region and hence behaves essentially as if it

were unreflected), can be reduced to 1.25 or less with
changes in fuel cost and yield of less than 0.02
mill/’kwhr(e) and 0.2% per year respectively. The
enhanced neutron leakage from the core, which results
from the power flattening, is taken up by the fertile

blanket and does not represent a loss in breeding _

performance. | _ -
Second, attempts at power flattening in two dimen-

67

~ upward in one direction and concave downward in the

other, even though the integrated neutron current over
the entire boundary of the central zone vanishes. In
view of these tendencies, it may be anticipated that a
flattened power distribution would be difficult to
maintain if graphite dimensional changes, resulting from
exposure to fast neutrons, were allowed to influence
the salt volume fractions very strongly. Consequently, a

core of the design shown in Fig. 5.4 was under

consideration as a means of reducing the sensitivity of

~ the power distribution to graphite dimensional changes.

6.4 Fuel Cell Calculations

A series of calculations was performed to investigate

| the nuclear characteristics of the two-fluid MSBR fuel

cells, or elements. These were based on the geometry
shown in Fig. 6.7. (Subsequent to these calculations, a
graphite sleeve was added around the fertile salt.)

‘The cell calculations were performed with the code
TONG and involved varying (1) cell diameter, (2) fuel
distribution (i.e., fuel separation distance), (3) 233U
concentration, (4) 222Th concentration, (5) fuel salt
volume fraction, and (6) fertile salt volume fraction.
Each of these parameters was varied separately while
holding the others constant. Figure 6.8 shows the effect
on reactivity of varying the parameters. The variations
are shown relative to a reference cell which had a

- diameter of 3 in., a fuel separation distance of %, in., a

fuel salt fraction of 0.1648, and a fertile salt fraction of
0.0585, with ~0.2 mole % 233UF, in the fuel salt and

27 mole % ThF, in the fertile salt.

sions have shown ‘that the power distribution is very

sensitive to details of composition and placement of the

flattened zone. Small differences in upper and lower

blanket composition, which are of no consequence in
the case of the uniform core, produce pronounced axial
asymmefry of the power distribution if too much axial
flattening is attempted. In addition, the axial and radial
buckled zones may interact through the flattened zone

to some extent, giving a distribution that is concave

These calculations showed that as the cell diameter
increases, the increased self-shielding of the 232Th
resonances leads to an increase in the reactivity of the
cell. Thus a decrease in breeding ratio associated with
the decreased >?2Th resonance integral is accompanied

ORNL DWG 70-2175

FERTILE SALT.

FUEL SALT

   

GRAPHITE =~ _
- 4 FUEL SEPARATION
DISTANCE.

Fig. 6.7. Geometry Used in Fuel Cell Calculations.

 
 

 

 

ORNL-DWG 67-4790

PERCENT FUEL SEPARATION CHANGE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

' -100 0 100 - 200
0.45 T
0.10.
1Y /
~ Q.05 - = R
|’ % cw- DIAMETER L —
- FUEL | —1
2‘ O . — I
//1/0\* / %,
N ) .
-0.05 $V°% &CO/IM
. <7 \'64/
& / 2
\9/ N
L "/ O
-040 — T
al
-40 -20 0 20 - 40 60

PERCENT CHANGE

Fig. 6.8. Effect on Reactivity of Changmg Cell Properties. Changes in all parameters are shown at the bottom for the fuel

separation that is shown at the top.

by a decrease in the required 223U loading. Optimiza-
tion calculations using cross sections based on 3- and

S-in-diam cells indicated that the annual fuel yield of

the system is essentially insensitive to fuel cell diam-
eters between 3 and 5 in, This is significant because the
larger cells are preferred for hydrodynamic reasons,

particularly in order to achieve the desired Reynolds

numbers for the fuel salt flow in the channels,

Table 6.7 and Fig. 6.9 show the flux distribution in
the S5-in.-diam- cell. Table 6.7 gives the ratio of the
average flux in the fuel to the cell average flux, the ratio
of the average flux in the graphite to the cell average
flux, and the ratio of the average flux in the fertile salt
to the cell average flux for the epithermal and fast flux
ranges. Figure 6.9 shows the thermal flux distribution
in the cell.

Two-dimensional diffusion-theory calculations indi-
cated that the central cell of the reactor may be useful

for control purposes. For example, if the central cell is

a 5-in-0D X 4-in.-ID graphite tube and if this com-
pletely empty tube is filled with fertile salt, the change

~ in reactivity is 8k/k = —0.018%. If the empty tube is

filled with graphite, the reactivity change is 8k/k =
+0.0012%. Thus there appears to be a substantial
amount of reactivity control available by varying the
height of the fertile column in the tubes, which might

be accomplished through use of a movable graphite

plug.

_ Table 6.7. Flux Ratios in Epithermal

 

 

 

and Fast Energy Ranges
; _ Ratio of Average Flux

Fuel - Graphite Fertile
0.821-10Mev  1.226 0929 0878
0.0318-0.821 Mev 1.090 0984 0.958
1.234-31.82 kev 1.014 - 0.998 0.991
0.0479-1.234 kev - 1.0 1.0 1.0
1.86-47.9 ev ‘ 1.0 : 1.0 1.0

 

6.5 Temperature Coefficients of Reactivity

In analyzing power transients in the two-fluid MSBR,

one must be able to determine the reactivity effects of

temperature changes in the fuel salt, the fertile salt, and
the graphite moderator. Since the fuel is also the
coolant and since only small fractions of the total heat
are generated in the fertile salt and in the moderator,
one expects very much smaller temperature changes in

the latter components than in the fuel during a power
transient. Expansion of the fuel salt, which removes -
“fuel from the active core, is thus the principal inherent

mechanism for compensating any reactivity additions.

We accordingly calculated the magnitudes of the
temperature coefficients of reactivity separately for the
fuel salt, the fertile salt, and the graphite over the range
 

 

  

2.0 -

o

FUEL —-|+/ A.I-

GRAPHITE

FLUX (arbitrary units)
o

o
o

0 { 2 3 4

FUEL

69

ORNL—DWG 67-479¢

CELL AVERAGE FLUX

GRAPHITE

: FERTILE

5 6 7 8 9

RADIAL POSITION {cm)

Fig. 6.9. Thermal Flux Distribution inCelt, E<1.86 ev.

of temperatures from 800 to 1000°K. The results of
these calculations, as shown in Fig. 6.10g, illustrate the
change in multiplication factor vs moderator tempera-
ture (with 8k arbitrarily set equal to zero at 900°K).
Similar curves of 6k vs temperature for fuel and fertile
salts are shown in Figs. 6.10b and 6.10c, and the
combined effects are shown in Fig. 6.10d. All these
curves are nearly linear, the slopes being the tempera-
ture coefficients of reactivity. The magnitudes of the
coefficients at 900°K are shown in Table 6.8.

The moderator coefficient comes almost entirely
from changes in the spectrum-averaged cross sections. It
is particularly worthy of note that the moderator
coefficient appears to be quite insensitive to uncer-
tainties in the energy dependence of the 233U cross
_sections in the energy range below 1 ev. This is to say
that reasonable choices of cross sections based on
- available experimental data yield essentially the same
coefficient. - | -

The fertile salt reactivity coefficient comprises a
strong positive component due to salt expansion (and
hence reduction in the number of fertile atoms per unit
core volume) and an appreciable negative component
due to temperature dependence of the effective
resonance-absorption cross sections, so that the overall

coefficient, though positive, is less than half as large as-

- that due to salt expansion alone, R
The fuel salt coefficient is due mainly to expansion of

the salt, which of course reduces the average density of -

~fuel in the core. Even if all core components were to
undergo equal temperature changes, the fuel salt coeffi-
cient would dominate. In transients in which the fuel

ORNL-DWG 67-11812

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 .

MODERATOR "FERTILE SALT
w4 : -
o . .

x - , / ,
& : '
i) 18
-4 : :
8 * |
 FUEL SALT ~ OVERALL
° - N
-
o
X 9
* .
0
-4
o | ||
-8 — . . -
800 900 1000 800 900 1000
| ' T (°K)

TCK)
Fig. 76.10._ MSBR 'Mfiltifih‘catipn Factor vs Temperature,
temperature change is far larger than that of the other

components, the fuel coefficient is even more con-
trolling.

 
 

 

Table 6.8. Temperature
Coefficients of '
Reactivity

 

Coefficient

Component ’tglzc' [ (‘_’K)-l]

 

| X 1075

Moderator +1.66
Fertile salt +2.05

. Fuel salt -8.05

" QOverall -4.34

 

6.6 Dynamics Analysis

The dynamic behavior of the MSBR, particularly the
reactor stability, was investigated using a linearized

- model of the two-fluid system. The model included a

lumped parameter representation of the neutronics
(including pure time delays for out-of-core precursor
transport), fuel salt heat transfer in the core, fertile salt

. heat transfer in the core, fuel salt heat exchanger, and

the salt side of the boiler and reheater. The heat
removal from the boiler and .reheater was assumed
constant. The resulting model consisted of 34 coupled
differential equations with 15 pure time delays.

The estimates of the temperature coefficients of
reactivity for the fuel salt, fertile salt, and graphite were
revised during the course of these dynamics calcula-
tions. Some of the calculations were based on the early
values and some were based on the later ones. Both

values are shown in 'I‘able 6.9. The neutron generation

time was 3.3 X 10™* sec.

The model was used for analyses of system stability,
transient response, and frequency response. The sta-
bility analysis (using the newer temperature coefficients
in Table 6.9) was accomplished by employing the
modified Mikhailov method described by Wright.*! The
analysis showed that the system is linearly stable.

The stability of the system is also indicated by the
response of the system to step changes in reactivity.
The linearized response of the reactor power to a step

~ change of 10™* Ak/k is shown in Fig. 6.11. This curve

is based on the old reactivity coefficients given in

- Table 6.9. Since -the model is linear, the response to

some other reactivity step is the product of the
computed response and the ratio of the new reactivity

 

41y, C Wright, An Efficient, Computer-Oriented Method for .
Stability Analysis of Very Large Systems, dissertation com-

pleted at the University of Tennessee, June 1968.

70

Table 6.9. Reactivity Coefficients

 

 

 

(Ak/k per °F)

_0l1d Value New Value
Fuel salt —4.6X 1075 —4.54 %1075
Fertile salt +1.43 X 107% +1.12X 107
Graphite +5,1X10°° +9.2%107¢
30 ORNL-DWG T70-2176

POWER CHANGE [Mw (1]

 

O 40 - 80 120 160 200
' “TIME (sec) :

Fig. 6.11. Power Trans:ent Following a Reactmty Step ot‘
107% sk/k with Reactor Operating at 556 Mw(t).

to the old reactivity. The linear results are obviously
not valid for large reactivity inputs, but would be
sufficiently -accurate for transients in which the power

~ changes by less than 10%. The response shown in Fig.

6.11 is expressed as the deviation from the full-power
output [556 Mw(t)] of a single reactor module. -

The power-to-reactivity frequency response of the
reactor is shown in Figs. 6.12 and 6.13 for the case of
full-power operation. In this instarice the results are
based on the newer reactivity coefficients given in
Table 6.9. As would be expected from the transient re-
sponse results, there are no tall peaks in the frequency
response amplitude which would indicate strong reso-
nance behavior. The frequency response was also com-
puted using the old reactivity coefficients. Since the

change in the results was very small, the transient .

response calculations were not repeated.
 

ul

71

ORNL-DWG 70-2178

108

10%

GAIN (Mw (1) per 34/4)

 

103
103 10°2 1o~! 109 10 102
FREQUENCY (radians/sec)

Fig. 6.12. Amplitude of the Power-to-Reactivity Frequency
Response. Reactor at full power,

120 -bWG T0-2177

80

40

PHASE ANGLE (deg)

 

1073 1072 1071 . 10° 10! 102
" FREQUENCY (radians /sec) :

Fig. 6.13. Phase of the Power-to-Reachvxty Frequency
Response. Reactor at full power,

In general, the system is well behaved dynamically,

~ and satisfactory operation should not be difficult to

obtain.

7. COST ESTIMATES
71 Gene:_al. |

One of the promising aspects of the molten-salt
breeder reactor is the potential for producing low-cost
power. At the present stage of development, accurate
detailed cost estimates are not possible, but our best.
estimate of the construction cost of a two-fluid
1000-Mw(e) MSBR station is about $140/kw(e). This
estimate is in terms of early 1968 conditions and value
of the dollar, and includes indirect costs. The estimated

net cost to produce power with private ownership of
the plant is about 4 mills/kwhr.

In making the cost estimates we assumed that an
established molten-salt reactor industry exists and that
materials are being supplied and plants are being
constructed and licensed on a routine basis. We also
assumed that the indirect charges, or owner’s costs, for
a molten-salt reactor are not significantly different from
those for other types of reactors.

Although the chemical reprocessing plant is part of
the reactor station, not all the chemical plant costs are . -
included in the estimate of the station construction
cost. The cost of the shielded cells to house the
chemical plant is included in the overall structures
account for the reactor plant, but the cost of the
processing equipment, the fuel and blanket salt inven-
tories, and the operation of the chemical plant was kept
separate from the rest of the station costs in order to
arrive at a fuel cycle cost which is comparable with the
fuel cycle- costs for other types of nuclear power
stations. The estimated fuel cycle cost is about 0.5
mill/kwhr, or about 0.7 mill/kwhr if the expense of
periodic replacement of the reactor vessels and graphite

‘cores is included. This is lower than has been projected
for most other types of nuclear power plants and

accounts for much of the interest in molten-salt
reactors.

The costs reported here for the tWO-flUId MSBR are
higher than those published in ORNL-3996' in 1966.
This is primarily because of changes in the plant

" concept, modifications to the design due to revisions in

the physical properties data, and escalation of costs
between 1966 and 1968. The present estimate of the
direct construction cost is. about the same as the
estimated cost for a pressurized-water reactor of
1000-Mw(e) size built on the same site. Because the
accuracy of the estimates is uncertain, we think the
major value is in comparing MSBR and PWR costs to
learn where the inherent differences in the systems have
an important bearing on the relative costs of the two
kinds of plants. Two areas stand out: The allowance for
maintenance is less on the PWR, but the cost of the
turbine-generator is less for the MSBR.

- 12 Conétruction Costs

The estimate of the construction cost of a two-fluid
1000-Mw(e) MSBR power station is summarized in
Table 7.1. Tables A.1 through A.11 in the appendix
give more -details of the costs. About half the total
construction cost is for conventional parts of the plant,
such as structures, turbine-generator, etc., for which

 
 

72

Table 7.1, Comparison of Construction Cost of Two-Fluid MSBR
and PWR 1000-Mw(e) Power Stations

 

 

 

 

 

 

Cost (millions of dollars)
MSBR? PWR?
Land (mcluded in indirect costs) ‘ _ -
Structures and improvements (see Table A, l) 10.6 146
Reactor equipment :
Reactor vessel (see Table A.2) 6.9 1.5
Graphite (sce Table A.3) 2.8 , -
_ Shielding and containment (see Table A 4) 5.1 c
Heating and cooling systems 2.1 39
Cranes 02 - 0.2
Control rods - 1.0 24
Heat transfer systems (see Table A.5) 211 19.7
Drain tanks or nuclear fuel handling (see Table A.6) - 4.2 1.8
Waste treatment and disposal 0.5 0.5
Instrumentation and controls : 4.1 -39
Feedwater supply and treatment (see Table A.7) 4.8 3.7
Steam piping 4.8 54
Maintenance equipment allowance 5.0 0.9
Turbine-generator (see Table A.8) 232 36.1
Accessory electrical (see Table A.9) 4.5 4.8
Miscellaneous (sce Table A.10) 1.6 1.3
~ Total direct construction cost 102.5 106.7
Sales tax and indirect costs (see Table A.11)4 384 40.0
Total construction cost

140.9 146.7

 

4MSBR costs are in early 1968 value of the dollar.

DPWR costs taken from J. A, Lane, M. L. Myers, and R, C. Olson, Power Plant Capital Cost
Normalization, ORNL-TM-2385 (June 1969), plus 4% escalation for the 1968 dollar. Since the
costs in the Lane study were based on the 1967 value of the dollar, they were escalated by 4% to
more closely approximate 1968 conditions. The PWR indirect costs used in the study were not
used in the two-fluid MSBR estimate shown in Table 7.1 because they were not on the same basis.
For simplicity, the same value for indirect costs of 33.5% was applied to the direct construction
cost of both the MSBR and PWR. A sales tax of 3% was added to the construction cost of both

types of plants.

‘PWR shielding cost included in structures and improvements,
dAssumes 3% sales tax and MSBR indirect costs 33.5% as shown in Table A,11.

costs are relatively well established. The reactor-
associated costs are less certain because of the pre-
liminary nature of the designs and the use of special
graphite and Hastelloy N for which there is no
experience in large-scale production and fabrication.
With regard to the graphite, a long-term cost of $5/1b
(see Table A.3) was used in these estimates. More recent

studies by Cook er al.*? suggest that the price could

approach $8/Ib. Installed tosts of Hastelloy N com-

 

42w H. Cook, W. P. Eatherly, and H. E. McCoy, Estimate of
Core Graphite Cost, ORNL internal correspondence
MSR-68:150 (Nov. 1, 1968). '

ponents were assumed to vary between $8 and $20/Ib,
depending upon the form of the Hastelloy N and
complexity of construction, as shown in Table A.2.

A few of the items listed in Table 7.1, such as

maintenance equipment, are subject to considerable -

uncertainty because little design work was completed in
those areas. A study by Blumberg® indicated that
about $5 million should be allowed for mamtenance
equxpment for the MSBR station.

 

43R, Blumberg, Preliminary Cost Estimate for Remote
Maintenance of the MSBR, ORNL internal correspondence

MSR-68-140 (Oct. 14, 1968).
 

 

 

[

"The comparative costs shown in Table 7.1 for a PWR
station were taken from the normalization studies by
Lane ef al.4* The MSBR and PWR costs are about equal
in many areas, but in at least two instances the
differences are worthy of note:

1. The maintenance equipment required for replace-
ment of the reactor vessels in the MSBR station is
included as a capital expense. This allowance is con-

siderably higher than corresponding PWR costs. The

replacement costs in terms of the materials and special
labor required were handled separately, as discussed in
Sect. 7.3.. :

2. The cost of the turbme-generator and assocmted
turbine plant equipment is much less for the MSBR
station because it operates at about 45% thermal
efficiency, as compared with the 33% thermal effi-
ciency of the PWR, and it uses supercritical-pressure

steam rather than the low-pressure steam of the water

reactor plant.

7 3 Power Productlon Costs

The total cost to produce electric power in a pnvate]y
owned - 1000Mw(e) two-fluid MSBR station is esti-

mated to be about 4 mills/fkwhr. The costs are

summarized in Table 7.2. The fuel -cycle cost ‘is
sufficiently low that even the addition of the expense
of periodically replacing reactor vessels and graphite

- results in a combined cost of only about 0.7 mill/kwhr.

The net cost of about 4 mills/kwhr to produce
electricity is attractively low.

The fixed charge of 13.7% used in making the
estimates is explained in Table A.12 in the appendix. In
estimating the depreciation allowance, a 30-year plant
life was assumed in order to be consistent with other
reactor evaluation studies. The possibility that the low
fuel cycle cost and higher thermal efficiency would

" make the useful life of an MSBR considerably greater

than 30 years was considered. An increase in plant life
to 45 years would produce a net reduction in the power

 

43PWR costs taken from J. A. Lane, M, L. Myers, and R. C.
Olson, Power Plant Capital Cost Normalization, ORNL-TM-
2385 (June 1969), plus 4% escalation for the 1968 dollar. Since
the costs in the Lane study were based on the 1967 value of the
dollar, they were escalated by 4% to more closely approximate

.+ 1968 conditions. The PWR indirect costs used in the study were
not used in the two-fluid MSBR estimate shown in Table 7.1

because they were not on the same basis. For simplicity, the
same value for indirect costs of 33.5% was applied to the direct
construction cost of both the MSBR and PWR. A sales tax of
3% was added to the construction cost of both types of plants.

Ta!ile 7.2. Estimated Electric Power Production Costs for a
Privately Owned Two-Fluid MSBR 1000-Mw(e) Power Station

 

In millsfkwhr
Capital cost? =~ . : 2.8
Fuel cycle cost
Reactor vessel and graphite rcplacement 0.2
cost (see Table A.13) -
Chemical reprocessing cost (see Tables 0.5
6.1 and 6.4) _
Operating cost (see Table A.14) 0.3
Total _ , : 3.8

 

8Capital cost based on total construction cost shown in Table
7.1, on fixed charges of 13.7% per annum, as listed in Table
'A.12, and a plant factor of 80%.

production cost of a little less than 0.1 mill/kwhr, as
explained in footnote b of Table A.12.

In applying the fixed charge, no distinction was made
between depreciating and nondepreciating capital in-
vestment except for the inventory components of the
fuel cycle cost.. Such a refinement to the estimate
would be overshadowed by uncertainties in other costs.
The salvage value of many of the items is not clear; the

- costs of decontaminating and reclaiming such things as

land, salt inventory, etc., must be balanced against the
intrinsic worth and the expense of disposal that would
otherwise be requlred

The estimate of the cost of replacmg the reactor
vessels at the end of the useful life of the graphite cores
is summarized in Table A.13. As explained in the

- footnotes to the table, an indirect cost of 10% was

applied to the procurement of the replacement reactors.
Many of the indirect costs of first construction would
not be applicable to the replacement equipment. The
lifetime of the core is such that the replacements can be
made during periods of extensive general maintenance
of the plant and turbine generators, so no outage other
than that included in the 80% plant factor was charged
against the production -cost. This seemed to be a
reasonable approach since no downtime is required for
refueling. If additional time were required for replacing
the reactor vessels, .the cost of power would be
increased by about 0.05 mfl]lkwhr for each two weeks
of extra time.? |

- Labor costs in addltlon to those for the regular plant
“maintenance crew were included in the reactor replace-

s

 

- 45Roy C. Robertson, Effect of Core Graphite Life on Power

Production Costs in Two-Fluld and Single-Fluid Molten-Salt -
Breeder Reactors in 1000-Mw(e] Power Stations, ORNL.
internal correspondence MSR-68-46 (Mar, 4, 1968).

 
 

 

ment expenses. Allowance was made for a special crew
of 18 men at $10/hr on a three-shift basis over a
two-month period. This time would include preparatory
and cleanup operations. The total cost of the special
labor amounts to $300,000. Scheduling of the replace-
ment and other maintenance operations was not con-
sidered in detail. The simplifying assumption was made
that the four modules would be replaced every elght
years.

A capital cost associated with replacement of reactor
vessels and cores was obtained by use of a replacement

- cost factor. The capital needed at the present time to

amount to $1 eight years hence is $0.63, 16 years hence
is $0.39, and 24 years hence is $0.25 if the interest rate
is 6%. The total replacement cost factor is the sum of
these, or 1.27. The total replacement cost of four
vessels and cores is $11 million, so the capital that must
be set aside at the time of plant construction for future
reactor replacements is $14 million. This capital would
not incur all the fixed charges given in Table A.12. A
rate of 8% was taken as being more appropriate in
arriving at a total replacement cost of about 0.2
mili/kwhr.

The estimated fuel cycle cost for the two-fluid MSBR

is summarized in Table 6.4. Among the major con-

stituents of the cost are items associated with the large
capital investment in inventories of fissile and fertile

materials and carrier salt. The inventories were treated

74

as a nondepreciating investment subject to fixed charges
of 10%. The daily makeup and discard of salt in the
processing plant amount to complete replacement of
the fuel carrier salt every fi five years. . )

The cost associated with the investment in processmg
equipment is about 0.1 mill/kwhr and is based on the
equipment costs reported in ORNL-3996.! This cost,
escalated to 1968 conditions, as explained in Table 6.4,
is included in the fuel cycle cost. The usual fixed

- charges of 13.7% and 80% plant factor were applied.

' Operating costs associated only with the chemical
processing were also included. A product credit of
about 0.1 mill/kwhr was estimated on the basis of a 4%
yield and 2?*UF, worth of about $14/g. The total fuel
cycle cost is about 0.5 mill/kwhr, or about 0.7
mill/kwhr if the expense of replacmg the reactor vessels
and cores is included. '

The costs for operating the power statlon are sum-
marized in Table A.14. The total is about 0.3
mill/kwhr. It includes labor and materials for normal
operation and maintenance, insurance, and miscella-
neous services. Also included is the expense of replacing
coolant salt. This estimate assumes 2% makeup per year

and a cost of about 25¢/lb for sodium fluoroborate.

Subsequent study and allowances for escalation have
resulted in more recent estimates of 50¢/Ib, although
this would possibly be reduced by quantlty buying in
an MSBR industry. '
 

 

 

 

b

&

'APPENDIX A:

Table A.1. Estimated Cost of Improvements, Buildings, and

COST ESTIMATES

Table A.2. Estimated Cost of Four Reactor Vessels for a

 

 

 

 

 

 

 

Structures for a 1000-Mw(e) Power Station. 1000-Mw(e) Power Station
Cost Dimensions
(thousands In feet
of dollars)  yooel diameter® 134
Ground improvements 800 Vessel height? o 17
Buildings and structures : Weights per Module
" Reactor building (see Fig. 4.4) . In pounds '
Excavation, 10,000 yd? at $8/yd> 80 Hastelloy N at $8/b .
Substructure concrete, 8600 yd® at $120/yd® 1,032 s 2894
Above-grade concrete, 11,120 yd at $80/yd3 890 Tangent 1’044
Confinement building, 2.3 X 10° ft* at $1/ft> 2,304 Wall liner 9.984
Turbine building, 290 X 115 X 125 at $0.60/ft> 2,501 * Cylinder 43,902
Feedwater heater space, ' 696 . 57,824
50 X 290 X 80 at $0.60/ft3 _ Hastelloy N at $10/1b
Offices, 50 X 240 X 20 at $1.50/ft3 355 . Baserings, Ib 7,855
Control rooms, 50 X 165 X 20 at $1.50/ft3 248 Hastelloy N at $12/Ib
Shop - - 555 Bottom outside head, Ib 20,053
50 X 165 X 80+ 50 X 265 X 20 at $0.60/£t° | Top outside head, Ib ;Z:‘s‘:z
Waste disposal building - | . : 150 Hastelloy N at $20/ib :
Stack 200 Bottom inside head, Ib 7,844
Warehouse 40 Top inside head, Ib 17,361
Intake screen structure for cooling water 700 Bottom deflector, 1b 3,417
Miscellaneous ‘ : 30 28,622
Total 10,600 ‘Costs
' In millions of dollars
Hastelloy N at $8/Ib 0.463
Hastelloy N at $10/1b - 0,079
Hastelloy N at $12/1b 0.451
Hastelloy N at $20/1b 0.572
Total for one module 1.565
~ Total + 10% contingency 1.722

75

Total for four modules .

 

4Dimensions of the core used in the cost estimates are shown

in Table A.3.
 

 

 

Table A.3. Weights and Esfimated Costs of Graphite for

76

Table A.5. Estimated Cost of Heat Transfer Equipment

 

 

 

 

 

 

@The ¥ ¢-in. stainless steel liner used in the cost estimates in -
- ORNL-MSR-6846 (zef. 45) is now judged too thm and % in. is

used here,

" DThe insulation cost of $1/ft? used in ORNL-MSR-68-46 (ref .

45) is behe_ved to be low. Although still not known with any
certainty the $2/ft? cost used here is probably more realistic.
Some believe that the cost of the insulation would be even
higher.

Four Reactor Modules for a 1000-Mw(e) Power Station
‘Com : - Cost
.Diameter, ft ' 100 (thousands
Height, ft » : 13.3 of dollars)
Volume, ft3/module 1041 - ‘
Volume-fraction graphite _ 4 0.80 Fuel salt primary heat exchange system
Volume graphite, ft3/module 830 Primary heat exchanger (4)
Volume graphite, ft® for four modules - 3320 12,530 X 4 X $103/f¢ 5,16_2
* Blanket and reflector o . Hangers, etc. 300
Outer diameter, ft _ 13.4 Fuel salt pumps (4) _ o
Height, ft 17 . Bowls » - 300
Volume blanket and reflector, ft3[module : - 2397 Pumps _ 700
| Volume-fraction graphite 0.42 Piping 200
" Volume blanket and reflector g:aphlte, {13 /module 423 ' '
Volume blanket and reflector graphite, 1692 Blanket salt primary heat exchange system
ft3 for four modules Primary heat exchanger (4)
3 , 1318 ft2 X 4 X $190/ft2 : - 1,000
Total graphite in four modules, fi - 5012 Hangers, etc. _ 200
Totg§ weight of grapl;ite in four modules, 561.3 Blanket salt pumps (4) '
Total cost of graphite in four modules, $ 10_6 | 2.81 Pumps , S | 300
(at $5/1b) S Piping - ‘ 100
Coolant salt circulating system _
Pumps (4) 2,000
Piping - ' ' 1,000
. ' . yae . Steam generators (16) '
Table A4. Estimated Cost of Shielding and Containment 2 2
" for a 1000-Mw(e) MSBR Power Station s 2915 f; X 16 ’; $140/1t . 6,530
Note: This account covers the cost of the thermal shields t;grznsr:tzea;(tears)g $) 130/ '2.832
for the cell walls, The cost of the concrete biological Coolant sal Iv and tr @ ’3 00
shielding and the confinement bulldlng is mcluded in 00 _ t supply and treatmen : _
Table 7.1. Total , 21,124
3-in. carbon steel plate 9Does not include coolant salt drain tanks (see Table A.6)._
1,861,000 Ib/module X 4 X $0.35/Ib $2,605,000 ' :
3/15-in; carbon steel plate - ‘
57,000 Ib/module X 4 X $0.50/1b 114,000
lé-in. carbon steel plate ,
137,800 Ib/module X 4 X $0.35/1b 193,000
316,600 Ib/module X.2 X $0.35/Ib 222,000
1, -in. stainless steel liner?
70,200 Ib/module X 4 X $1.50/Ib 422,000
56,6001b X 2 X $l.50/lb 170,000
Insulation? '
127,000 ft2 /module X 4 X $2/ft2 1,016,000
100,000 £t2 X 2 X $2/ft2 400,000
~ Total ' $5,142,000
o

ol

Table A.6. Estimated Drain Tank Costs

77

 

Table A.S. Estimatéd Turbhe-Geflmtm Piant Costs for a2

 

 

 

 

 

 

 

 

 

 

1000-Mw(e) Power Station
Fuel Salt Drain Tanks Turbine-generator unit? -$18,970,000
Volume of salt stored, ft” 1444 Circulating water system® 1,460,000
Storage capacity per tank, ft 1200 Condenser and auxiliaries 1,700,000
- Number of tanks (2 per module) 8 Central lube oil system 80,000
Inside diameter, in.% 48  Turbine plant instrumentation 400,000
Weight _ ~ Turbine plant piping 220,000
Shell and heads, 16 82,000 Aucxiliary equipment for generator 75,000
c Tltxbes and tube sheets, Ib 115,000 Other turbine plant equipment 125,000
os ' - Turbine bypass (25% throttle flow 300,000
- Shell and heads at $8/lb $106 0.7 ypass (25% ) —
Tubes and tube sheets at $20/1b, $10° 2.3 Total $23,200,000
Heat removal system allowance, $10 0.5 ABased on 3600- q .
Total for fuel salt tanks, $10° 3.5 yosed on 3600-1pm tandem-compound unt, ;
' Condensing water intake structure and screens are included
Blanket Salt Drain Tanks with structures and improvements (Table A.1).
Volume of salt stored, 3 2500 ,
Number of tanks (4 per module) 16
Inside diameter, in. : 12
Height, ft 20
Wall thickness, in. 0.5
‘Weight, totallb 42,900
Cost at $10/1b + 15% allowance for nozzles, 0.3
etc., $10°
Coolant Salt Drain Tanks ‘
Volume of salt stored, ft> 1000 .
6 .
Cost allowance, $10 0.15 Table A.9. Accessory Electrical Costs for a
" Flush Salt Tanks 1000-Mw(e) Power Station
Cost allowancc SIO6 0.15
Switchgear $ 775,000
@Subsequent studies indicated that a 60-m.-dlam tank may Switchboards 285,000
be required. - Station service transformer 262,000
Auxiliary generator 78,000
Distributed items 3,100,000
Total $4,500,000
Table A.7. Estimated Cost of Feedwater Supply and
- Treatment System for a 1000-Mw(e) Station
o _ | Cost
- (thousands
_ of dollars) Table A.10. Miscellaneous Costs for a
Makeup water supply -. 4 1000-Mw(e) Power Station
Feedwater purification system 466 Turbine crane and hoists $ 300,000
_ Feedwater heaters - ' ._ _1299 ) Air and vacuum systems 300,000
Feedwater pumps and drives © 1600 - Communications systems 50.000
Reheat steam preheaters (8) 275 Machine tools ' 300’000
~ . Pressure-booster pumps (2) L 407 Service water ‘ - 300’000
. Total | | . 4050 Coolant salt inventory ($0.25/Ib) - 300,000
Total with 20% allowance for contingencies 4800 Total $1,600,000

 

 

 
 

i
i
|
i
1
:

 

- Table A.11., Explanation of Indu‘ect Costs

78

Table A.13. Graphite and Reactor Vessel Replacement Cost

 

 

 

Table A.12, Fixed Charge Rate Used for

 

 

Investor-Owned Power Station
Rate (%/year)

Return on money invested? 7.2
Thirty-year depreciationb 1.02
Interim replacements® 0.35
Federal income taxes? . 2.04
Other taxes® 2.84
Insurance other than liabilityf 0.25

Total 13.7

 

dReturn based on 52% in bonds at 4.615%

return, 48% in equity capital at 10%.

BThe sinking-fund method was used in deter-

mining the depreciation allowance for the 30-year
assumed life of the plant. The depreciation allow-
ance amounts to less than 8% of the fixed charges.
A 45-year life, say, would decrease this by about
two-thirds, and reduce the total fixed charges to
about 13.4% per annum.

€In accordance with FPC practlce, a 0.35%
allowance was made for replacement of equipment
having an anticipated life shorter than 30 years.

- (Reactor and graphite replacement is included in a

special operating cost account.)

j dFederal income taxes were based on “sum of
. the year digits” method of computing tax de-

ferrals. The sinking-fund method was used to

~ normalize this to a constant return per year.
€The recommended value of 2.84% was used for

other taxes.
‘FA conventional allowance of 0.25% was made

for property damage insurance. Third-party lia-

bility insurance is listed as an operating cost.

 

Used in Table 7.1 for 1000-Mw(e) Power Station
. Percent  Total Cost? Reactor vessel cost? $ 76X 10(i
. - - . Graphite cost? o _ 3.1x10°
General and administrative 4.7 $1.047 Power revenue loss® . : None
~ Miscellaneous construction 1.0 1.057 Labor costd ' 0.3 X 10°
Architect-engineer fees 5.1 1.111 - o e 6
Nuclear engincering fees 2.0 1.134 . Total for four modules per replacement $11.0 X 10"
Startup costs 0.7 . 1.142 Estimated life, years® '8
Contmgencsf _ 27 1.172 Replacement cost factor (see text) 127
. Interest during S-year construction 13.5 1.331 Thirty-year replacement cost ’ $14.0X 10°
period : _ Power production cost, mlllslkwhrf : 0.16
Land ($360,000) 1335 |
. : \ . 2Based on Table A.2 with 10% added for indirect costs. -
For direct cost of $1.. bBased on Table A.3 with 10% added for indirect costs.

CAssumes that reactor can be replaced within normal down-
time for plant and that no additional power outage is chargeable
to graphite replacement.

dLabor cost is in addition to that of operating crew.

¢Estimated life of graphite based on 20 kw/liter average core -
power density and a]lowable dose to graphite of 3 X 1022
neutrons/cm?,

fPower production cost for reactor replacement based on 8%
fixed charges for capital and 80% plant factor.

Table A.14. Operating Costs for a 1000-Mw(e) Power Station

 

 

Anmual
Cost
Total payroll, 70 employees with 20% $ 554,000
fringe benefits?
Private insurance 260,000
Federal insurance, at $30/Mw(t) 66,800
Repair and maintenance materials - 1,065,000

Makeup coolant salt, at 2% of capital cost 7,000
Contract services ' 71,500

Total annual operating cost® $2,024,300

 

2Does not include special crew used in replacing the reactor.
This special labor cost is included in the reactor replacement
cost shown in Table A.13,

bDoes not include materials for replacing the reactor vessel

‘and graphite (see Table A.13).

°Total operating cost in mills/kwhr based on 80% plant factor
is 0.29. This operating cost is essentially the same as that used
in other reactor evaluation studies.
T

o¥

WO NAN DA W -

Yot
=

11,
12.
13.
14,
15.

32,
33,
34,
35,
36
37,
38,
39,
40.

41.
42,
43.
44,
45.

. R.K. Adams

. G. M. Adamson
. R. G. Affel

. 1. L. Anderson

. R.F. Apple

. W, E. Atkinson
F. Baes

M. Baker

E. Bamberger
J. Barton

B. Bates

F. Bauman
E. Beall

R. Beatty

J. Bell
Bender

E. Bettis

S. Bettis

S. Billington
E. Blanco

F. Blankenship
O. Blomeke
Blumberg
G. Bohlmann
J. Borkowski
I. Bowers
E. Boyd
aunstein

. Bredig

. Briggs

. Bronstein
. Brunton
.Burke
antor

. Cardwell
. Carlsmith
. Carter

. Caton

. Cavin

S. L Chang

R. H. Chapman
C.J. Claffey
F. H. Clark
Nancy Cole

. C.
. L
. C.
. C
J.
H.
S.
H.
M.
. M
. C.
. E.
. D.
. R.
. F.
. .
. R.
. E.
. C.
. H.
. G.
. L
. M.
. R.
H.
G.
0.
S.
D.
R.
Ww.
J.
0.

wmfiwinécww>?

47.
48,
49.
50.
51.
52,
53.
54.
55.
- 56.
57.
58.
59.
-60.
61.
62.
63.

65.
66.
67.
68.
69.
70.
7.
72
73,

74,

75.
76.
T,
78.
)
80.

81.

82.

83.

85.
86.
87.
88.
89.

79

 

INTERNAL DISTRIBUTION

. Collins
. Compere
. Cook

. Dworkin
. Eatherly
R Engel
. I. Federer
Ferguson
Ferris

. E.

. M.

.P.
. K. Franzreb
. H.

. C.

K

.H.

. B. Gallaher

. E. Gehlbach .
. H. Gibbons

. 0. Gllpatnck
Goldberg |
. Grimes

Handley

. Heacker

. Haubenreich

91.

92.
03.

- 94,
95.
96.
97.
98,
. 99,

- 100.
101,

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112,

113.

114,

115.

116.
117. °

118.

119.

120.

121.

122
123.
124,
125.
126.
127.

128.
129,

“130.

131.

132.

133,

c:sncwmmwwzmwwhhe>w;9rm.

rmwvrflmorz—rw>m€m~w
=

FOw”Fz
st
5
2
&

W. R. Huntley
H. Inouye

W. H. Jordan
P. R. Xasten
R. J. Kedi
C.W.Kee

M. T. Kelley
M. J. Kelly

C. R. Kennedy
T. W. Kerlin
H.T. Kerr

Korsme)'rer
. Krakoviak

Krewson

(57

ane
ndauer

l-'r*;-_-
5

ndm

yon
acklin
acPherson

\on

=

len

zgz

acPherson

ORNL-4528
UC-80 — Reactor Technology

134,
135.

- 136.
137.
138.
139.

140.

141.
142.
143
144,
145.
146.
147.
148,
149.
150.
151.
152.
153
154,
155.
156.

157.
158.
1159.
160.

161.

163.
164.

165.

166.
167.
168.

169—178.
179.
180.
181.
182.
183.
184.
185,
186.

. K. McGlothlan
. J. McHargue
. A. McLain

Perry

Plper

. Ragan
. Redford

M. Richardson
Rlchardson _
. D. Robbins
.'C. Robertson
.G.Ross
R

P

. C. Savage
W. F. Schaffer
Dunlap Scott
J. L. Scott

H. E. Seagren

 
 

 

 

80

187.

C. E. Sessions - 202. E. H. Taylor 217. J. C. White

188. J. H. Schaffer 203, W. Terry _ 218. R. P. Wichner
189. W. H. Sides ‘ -~ 204. R.E. Thoma 219. L. V. Wilson
190. M. J. Skinner 205. L. M. Toth 220. Gale Young
191. G. M. Slaughter 206. D. B. Trauger 221. H. C. Young
192. A.N. Smith 207. Chia-Pao Tung 222. J.P. Young

- 193. F. J. Smith 208. W.E.Unger = ~ 223. E. L. Youngblood
194. G.P. Smith 209. G. M. Watson 224. F.C. Zapp
195. O. L. Smith 210. J. S. Watson 225-227. Central Research Library
196. I. Spiewak 211. H. L. Watts 228-229, ORNL — Y-12 Technical Library
197. H. H. Stone 212. C.F. Weaver Document Reference Section .
198. R. A. Strehlow 213. A. M. Weinberg 230—329. MSRP Director’s Office (Y-12)

- 199. R.D. Stulting - 214. 1.R. Weir 330—364. Laboratory Records Department
200. D. A. Sundberg 215. H. L. Whaley 365. Laboratory Records, ORNL R.C.
201. J. R, Tallackson 216. M. E. Whatley o

EXTERNAL DISTRIBUTION

366. Kermit Laughon, RDT Site Office, Oak Ridge National Laboratory
367. C. L. Matthews, RDT Site Office, Oak Ridge National Laboratory
368. D. F. Cope, RDT Site Office, Oak Ridge National Laboratory
369. T. W. McIntosh, Atomic Energy Commission, Washington
370. F. N. Peebles, University of Tennessee, Knoxville
371. H. M. Roth, Atomic Energy Commission, Oak Ridge
372. M. Shaw, Atomic Energy Commission, Washington
373. 1. A. Swartout, Union Carbide Corporation, New York
374. W. L. Smalley, Atomic Energy Commission, Oak Ridge
375. David Elias, Atomic Energy Commission, Washington
376. W. T. Hannum, Atomic Energy Commission, Washington
377. D. R. Riley, Atomic Energy Commission, Washington
378. 1.1J. Schreiber, Atomic Energy Commission, Washington
379. F. N. Watson, Atomic Energy Commission, Washington
380. M. J. Whitman, Atomic Energy Commission, Washington
381. Laboratory and University Division, AEC, ORO
382. AEC Patent Office, ORO
383—-590. Given d1str1but10n as shown in TID-4500 under Reactor Technology category (25 coples — CFSTI)