Le
$
1
!
.
.
| OF L
ORNLP
2178

-
.
:
1
.
.
.
,
..
PEETEEEE
i
后后后
​-
en
.
..'.
MICROCOPY RESOLUTION TEST CHART
NATIONAL BURN AU OF STANDARDS – 1963
.
om - -1278
Conf- 660524-
5
2178
CECIL PRICES
468/100 mo sa
MASTEX
JUN 27 366

THORIUM BREEDERS AND CONVERTERS: THEIR INFLUENCE
ON FUEL CYCLE ECONOMICS AND FUEL UTILIZATION
Paul R. Kasten

RELEASED FOR ANNOUNCEMENT
IN NUCLEAR SCIENCE ABSTRACTS
Presentation
at the
SECOND INTERNATIONAL THORIUM
FUEL CYCLE SYMPOSIUM
May 3-6, 1956
Gatlinburg, Tennessee
LEGAL NOTICE
This report no prepared uw account of Government sponsored work. Neither the United
Suntos, por the Commission, nor any person acung ca bebell of the Commission:
A. Kakos say w reaty or reprowatation, expressed or implied, mu respect to the accu-
racy, completeness, or wofaloes of the lalormation coaieload in the sport, or that the wo
of any information, appuntus, method, or process declosed in this report may not lofriage
printtoly owned righus; or
B. ASRING way liabilues with rospect to the use of, or for damagos remulung Irom the
Un ni uny laformation, appenatus, motbod, or procesi dieclosed in this report.
As used to the above, porno acting on behalf of the Commissioo" lacludes way om-
ploys or contractor of the Commission, or omployee of such coatructor, to an oxtont that
such employs or cuatrictor of the Commission, or employee of such contractor propurus,
dienomlastes, or provides access to, any information pursuant to die groplogueat or contract
wild the Commissivo, or to employment with such contractor.
OAK RIDGE NATIONAL LABORATORY
Oak Ridge, Tennessee
Operated by
UNION CARBIDE CORPORATION
for the
U.S. ATOMIC ENERGY COMMISSION
LUR

THORIUM BREIDERS AND CONVERTERS: THEIR INFLUENCE
ON FUEL CYCLE ECONOMICS AND FULL UTILIZATION*
Paul R. Kasten
Oak Ridge National Laboratory
Oak Ridge, Tennessee
Introduction
Important objectives of the commercial nuclear power program are to
develop reactors which produce low-cost power and at the same time conserve
nuclear fue) resources. Since the evaluation factor that encompasses all
others in commercial application of reactors 18 power production cost,
consideration of fuel utilizaiion aspects should be consistent with genera-
tion of low-cost power over a given period of time. Of course, in evaluating
economics, future conditions must be properly weighed and attitudes of
governmental agencies taken into consideration, Thus, thorium reactor devel-
opment should be such as to produce syntems which generate low-cost power
and at the same time give good nuclear fuel utilization. These character-
istics have to be at least as good, and probably better, than corresponding. :.
uranium systems in order to justify development, and the question arises: .'
Are thorium reactors better than uranium-fueled systems?
The answer to this question involves a number of facets which will be
discussed briefly. First of all, from the fertile material viewpoint, there
is ample supply of either uranium or thorium. Also, thoriram reactors make
use of a fertile material which in nature does not contain a fissile fuel.
Because of this, thorium systems are initially lependent upon lissile fuel
derived from natural uranium and in that sense are dependent upon the uranium
fuel cycle. At the same time it is well known that, in thermal reactors,
better neutron economy can be obtained using the thorium fuel. This advantage
18 accompanied by use of highly enriched fissile fuel and often by a total
fuel "enrichment" wich is higher than that of corresponding uranium-fueled
systens. These circumstances can lead to thorium fuel cycle costs which are ".
higher, lower, or about the same as those of corresponding uranium-based
cm
*Research sponsored by the U. s. Atomic Energy Commission under con- .
tract with the Union Carbide Corporation.
.
.. wet
Eystems, the particular situation changing with reactor type and techon-
logical status. Also, the physical and chemical properties of fuele can
have a significant effect on reactor performance through their influence
on fuel recycle costs and reactor operating conditions.
iii.
Results of Cost Evaluation Studies
!
The reactor types which presently have the most emphasis with regard
to use of thorium fuel cycles are: the high-temperature, gas-cooled,
graphite-moderated reactor (PGR); the beavy-water-moderated reactor (HWR,
with either beavy water, light water, or organic coolant); the seed-blanket,
light-water-cooled and moderated reactor (SBR); and the molten-salt-fueled,
graphite-moderated reactor (MSBR). The performance of these systems, along
with others, has been summarized by Rosenthal et al., with the results
obtained using the methods and procedures employed an the OANL evaluation of
advanced converters.?
The specific design characteristics of the above-mentioned concepts ...
are given by Rosenthal et al., and are reproduced in Table 1. Brief
descriptions oi the associated reactor core features are given below.
The HIG? is a helium-cooled, graphito- and BeO-moderated reactor based
on tue "TARGET" concept of General Atomics. Graphite fuel elements contain
loose pyrolytic carbon-coated fuel particles in holes located in a "phone-
dial" arrangement. Two sizes of particles are used, one containing only
uranium and the other only thorium. The two types or particles are sepa-
rated at the end of the cycle and processed separately. Material from the
urunium particle 18 sold or discarded to reduce the buildup of 4236 in the
system. Bred uranium recovered from the thorium particle 18 combined with
makeup U235 and refabricated into the urantım-only particles for a subse-
quent fuel cycle. Thus, the makeup U235 is never recycled and bred uranium
is recycled for only one pass. A Beo spine in the center of the graphite
body has a volume that gives a carbon-to-Be atomic ratio of 2.4.
The HWR concept is a pressure-tube type, heart-water-moderated and
cooled system based on a design study by E. I. du Pont de Nemours and Co.
(Savannah River Laboratory). Pressurized heavy water coolant passes through
688 Zr-2.5% NO pressure tubes which have an inside diameter of 4.43 in.
1
.
I
*-=
-
*
-
-
--
-
*
*
va
Table 1. Thorium Reactor Design Characteristics
High-
Temperature
Gas-Cooled
(BIGR)
Heavy Water
(HWR)
Seed
Blanket
(SAR)
Molten-
Salt
Breeder
(MSBR)
1010
3870
1000
2250
538/238
241/33
44
Net electical capacity, Mw(e)
Reactor power, Mw(th)
Steam temperature/pressure, °C/atm
Net station efficiency, %
Moderator
Coolant
Coolant temperature, inlet-
outlet, OC
Coolant pressure, atm
Control metuod
1008
2270
565/238
44
Graphite
Helium
382-800
30
Poison
rod
9.5 x 4.8
27
1000
3200
247/37
31
1.0
ico
271-206
140
Seed
position
5.3 x 2.2
Graphite
Fuel salt
566-732
.
26
D20
Dao
264-304
130
Continuous
fueling
7.9 x 4.6
16
Corcentric
tubes
1.34, 2.68,
3.83
UO2-Thoa
On-line
refueling
Continuous
fueling
2.6 x 2.6
...
Core diameter x beigat, meters
Specific power, kw/kg fertile
Core power density, kw/liter
Fuel elements
48
17-24
160
Molten salt
Graphite
cylinder
4.5
Fuel rod or tube diam, in.
68
ña
bundie
0.25, 0.33,
0.744
UO2-Thoa
Batch
--
Fuel
Fuel management
UC2-ThCa
1/12 scatter
refueling
UF4-LIF-Befa
Continuous
Circulated through external heat exchanger.
*****.com
and are 0.16 in. thick. Concentric tubular fuel elements are used, and
the reactor 18 refueled while at power.
The SBR 18 based on the movable-fuel concept proposed by Bettis
Atomic Power Laboratory. In this concept, fully enriched uranium 18 located
in annular seed regions distributed throughout a thorium blanket. The
SBR 18 controlled by axial movement of portions of the seed, which changes
the effective thickness of the seed annulus. This results in a change in
the leakage of neutrons from the highly reactive seed regions into the sub-
critical blanket regions, thereby providing criticality control. Control
poisons are thus eliminated, and the neutron economy improved. In the
"converter recycle" concept considered here, the seeds are initially fueled
with U235, and U233 18 recovered from the blanket. After about three con-
verter cycles using 1235 seeds, the accumulated u233 18 used to fuel the
core. According to BAPL, self-sustaining recycle is achieved afterwards
with no additional fuel makeup required.
The MSBR 18 a two-region, fluid-fuel concepi with fissile material in
the core stream and fertile material in the blanket streem. The fuel salt
18 in direct contact with graphite moderator, and graphite tubes are used
to separate core and blanket streams - The fertile stream not only surrounds
the core, forming a blanket region, but also circulates through che core
region. Fuel proceusing is accomplished in an on-site plant, utilizing
fluoride volatility and vacuum-distillation processing.
The ground rules and procedures followed in the evaluation of these
reactors are given in reference 1. In brief, the evaluation was based on
minimum power costs and associated nuclear performance, considering private
ownership of reactor plants, fuel, and fuel fabrication and processing
plants. Fuel recycle costs were based on single-purpose fabrication and
processing plants 'serving 15,000 Mwle) capacity of the cancept under con-
sideration. Fuel cycle costs were based on a 30-year reactor life, using
"levelized" costs ("present worth" method). The reactors were initially
fueled with U205 and Th, with recycle of bred fuel plus U235 as required.
Fuel cycle costs obtained are reproduced in Table 2.
Along with the fuel cycle costs, capital costs and operating and
maintenance costs were estimated. Insofar as possible, the cost estimates
Ata

Table 2. Thirty-Year-Average Fuel Cycle Costs for
Reference Economic Factors
HIGR
HWR
· SBR
MSBR
Equilibrium or last-cycle data
Exposure, Mwd/MT of fuel
52,300 29,600 12,000 --
Fuel lifetime, full-power years
4.95
2.18
Feed enrichment, wt% f18sile material 3.1
Fabrication and processing plant size for
15,000 MW(e), Mr of fuel per year
for
120
189 590 1,300C ():
Fabrication cost, $/kg of fuel : - 115.00 37.50 51.50
Processing cost, $/kg op fuel .. :' 105.00 39.70 23.00€
Fuel shipping costs (fabricated +
... 25.60 5.90 5.000
irradiated), $/ling of fuul
;
Cut-of-pile holdup time, days ..... 390 .330 290 --
Initial lissile inventory, ke. ... 2,910 1,450 3,500
Value of first core loading, $105 . 36.00 18.33 43.82
Fabrication cost of first core, $105 16.40 3.008.60 4.39%
Fuel cycle cost, mills/kwhrie).
Fabrication
0.26 0.22 0.61 0.088
Net fuel burnup and losses
0.19 0.40 0.12 -0.084
Processing
0.19 0.24 0.27 0.12
Shipping
0.05 0.04 0.05 0
Fissile and fertile inventory
0.51 0.43 0.95 0.17
Interest on fabrication
0.17 0.05 0.08 0.068
Interest on processing
-0.03 -0.01 -0.03 0 .
Net fuel cycle cost
1.34 1.39 2.05 0.35
620
7.44.
.
.
0.19
Losses = 1% in processing, +0.2% in fabrication.
'Includes cost of initial purchase of Beo.
For breeder; pre-breeder throughput is 860 MT/year.
For breeder; pre-breeder costs, $64/kg.
For breeder; pre-breeder costs, $30.50/kg. . .,
*Integrated processing.
costs associated with carrier salt, including; L17. :
Losses = 0.1% per pass through processing..
e
for the different reactor plants were on a comparable basis. The resulting -
power costs are reproduced in Table 3, along with the associated fuel con-
sumption characteristics. Also included for comparison purposes are the
results for a pressurized water reactor (PWR) and for an AWR using the uranium
fuel cycle. 1,2
Table 3. Power Cost of 1000-Mw(e) Plants and
Associated Nuclear Performance
(12% fixed charges on reactor plant and 10% fuel
inventory and fabrication interest charges)
Th Cycle
AWR SBR
U Cycle
MSBR PWR WR
HTGR
Power cost, mills/kwkr(ei
Capital
Operating
Fuel cycle
Heavy water
Total
1.9
0.3
1.3
2.2
0.3
0.4
2.1
0.3
1.6
2.1
0.3
2..2
2.2
0.3
1.4
0.5
44
0.90
0.22
2.1
0.3
2.0
-
44
1.00
0.00
0.5
Breeding ratio
Index of fuel consumption
3.5
0.84
0.60
2.9
1.06
-0.13
4.0
0.60
1.29
4.1
0.66
1.27
Cost of on-site processing facilities are included under "Pual cycle
cost od
°(1-BR)/thermal efficiency, a measure of relative fuel consumption.
As indicated in Table 3, the SB has better neutron economy but higher
fuel costs and power cost than does the PWR. The same appears true of the
HWR-Th versus the HWR-U.* Both the HTGR and the MSBR systems appear to
etter
*A significant improvement in plant thermal efficiency can be obtained
by considering organic coolant rather than heavy water coolant in the HWR
systems; this leads to significantly lower power costs and also to better
fuel utilization characteristics than given above for HWR!8. Although use
of organic coolant improves the competitive position of HWR power plants,
it appears that the HWR-Ta system will have higher costs than the HWR-U
system under conditions analogous to those considered in references 1 and 2.
have significantly lower power costs and better neutron economy than do
PWR systems, with the MSBR the only concept which operates with lowest
fuel cycle costs as a breeder. These results 11lustrate that there can
a significant variation in fuel cost cost and in power cost as the reactor
type 18 changed, and that use of the terrium cycle in itself does not .
insure a power cost lower than that of analogous uranium-fuel-cycle systems
or of PWR plants. At the same time, the results indicate that there are
thorium reactors which have low power costs and good fuel utilization
characteristics. In the following sections the ALTGR will be considered
as representative of a thorium converter system, and the MSBR as represen-
tative of a thort.um breeder system.
Basis for Thorium Use in HTGR and MSBR Systems
Molten-salt reactors make use of fluoride volatility processing to
recover valuable fissile material in an efficient and inexpensive operation.
Since ThF4 dots not form a volatile ThF6 compound, the fluoride volatility
processing step is uniquely suited to the thorium fuel cycle. In addition,
ThF4 dissolved in carrier salts does not undergo oxidation-reduction reactions
as does UF4, which reduces mass transfer effects in systems having high
fertile material concentrations and contained in Hastelloy N. Thus, the
chemical and physical properties of thorium and wanium fluorides are such
As to significantly favor use of the thorium cycle. In addition, even 11
U238 and Ta had identical physical and chemical properties, use of the
thorium fuel cycle probably gives fuel cycle costs slightly lower than
corresponding uranium-cycie fuel costs in these homogeneous-type (from
a nuclear viewpoint) reactors and also better fuel utilization character-
istics.
HIGR reactors can utilize either the uranium or the thorium fuel cycle,
but most of the emphasis has been placed on use si thorium. At least
partially, this choice was made initially to avoid the positive temperature
coefficient of reactivity 288ociated with buildup of bred plutoniwn; also,
use of the thorium fuel cycle gives higher conversion ratios. Actual i
comparisons of the two fuel cycles in HIGR systems, however, are rare; use
18 made here of the work of Carlsmith et al.3
.7,
ita:
-
.
.

more
. The;":
-
.
The reactor design, core parameters, economic ground rules and cost
bases used were similar to those given in reference 2, except that in the
study by Carlsmith et al. equilibrium conditions were used in evaluating
· the fuel recycle cases, and the reactors had lower fissile inventories
(the difference between "equilibrium" and "30-year life" Juel cycle coste
can be 0.2-0.35 mill/kwhr(e), with equilibrium conditions being more
favorable; lowering the P188ile inventory also led to more frequent fuel
movement). Thus, direct comparisons with previous results above should
not be made; however, the results were do indicate the relative performance
of thorium- and uranium-fueled HIIGR's, based on a system capacity of 15,000
Mwle). Remote fuel fabrication was considered for recycled fuel, hooded
fabrication for the thorium-U235 fuel, and direct fabrication for the
partially enriched uranium. The minimum fuel cycle costs obtained for the
different cases are given in Table 4, along with the fuel exposure and net
conversion ratio.
As shown, the fuel cycle costs are significantly lower for the thorlum
fuel cycle than for the uranium fuel cycle so long as the fuel is pid cessed
and bred fuel recovered. Also, it is disadvantageous to recycle bred fuel
in the uranium, while it is advantageous to do so in the thorium cycle
(only slightly lower fuel cycle cost but significantly higher conversion
ratio). The recycled thurium-uranium fuel has the lowest fuel cycle cost
in spite of relatively high costs associated with remote funl fabrication.
Table 4 also gives the fuel cycle costs if the exposed fues were not
processed, but discarded. Under this situation it is less expensive to
use the uranium fuel cycle, which indicates the need for a recycle industry
when utilizing the thorium fuel cycle.
The above study also indicated that under equilibrium conditions the
conversion ratio is about 0.95 in recycle thorium HTGR's if exposures are
about 25-30,000 Mwd/T, and that breeding ratios of unity can be obtained
for short fuel exposures. Thus, from the viewpoint of fuel cycle cost and
1mproved fuel utilization, it is important that means be found for reducing
the fabrication and processing costs associated with HUGR fuels.
Table 4. Minimum Fuel Cycle Costs for HIGR Systems Using
Either tire Thorium Fuel Cycle or the Uranium Fuel Cycle
Minimum Friel Cycle Costs, Mills/kwhrie)
Tunium Fuel Cycle
Thorium Fuel Cycle
Breui Fuel
Not
Recycled
Bred Fuel
Recycled
(equil. cond.)
Bred Fuel
Not
Recycled
Bred Fre).
Recycled
(equii. cond.)
0.97
1.20
:
0.86
0.88
Total fuel cycle cost (with
processing), mills/kwar(e)
Fuel cycle cost without
. processing, and discard
spent fuel, mills/kwhr(e)
Fuel exposure,* MWD/T
0.93
1.01
102,000
0.63
59,000
0.63
87, ovo
0.72
50,000
0.86
Net conversion ratio*
**Based on cycle with recovery of bred fuel.
.
.
E
.
20
Value of U233 Availability
The above results for HIGR systems indicate that a significant advan.
tage 18 associated with U233 availability. However, because of its value
in recycle, the only economic way to produce "extra" u233 18 to develop
an economic breeder operating on the thorium fuel cycle. The MSBR offers
such a possibility, so that it 18 realistic to consider that umas can be
bought commercially. Whether this lissile fuel should be 18 ed in starting
up additional MSBR'S, HIOR's, or fueling existing HIGR' 8 18 dependent upon
the specific economic situation. A recent study on the fuel value of
ves 10 HIGR systems gives some information on this question.
In the referenced study, the HIGR reactor design and economic bases
were essentially those used in the advanced converter evaluation reports
considering four fuel recycle modes; these were:
1. Non-recycl.e: The reactor 18 fed with U235 and the discharged uranium
18 sold. Makeup U235 18 kept separate from thorium and hred uranium.
2. Full recycle: All the recovered uranium 18 recycled with U235 makeup.
Uranium and thorium are mixed together in one particle.
3. Type I segregation: Only bred uranium from the tiborium particle is
recycled. The recycle uranium and makeup U235 are combined in a
particle separate from the trium particle. Material recovered from
this uranium particle following irradiation 18 sold or discarded and
not recycled.
Type II regregation: Makeup U235 18 kept separate from thorium and
recycle uranium, and its residue 18 not recycled. Bred uranium 18
recycled back into the thorium particle, and no bred material is sold.
The value analysis for the HTGR was performed by comparing 30-year
average costs for the different recycle modes. The full recycle and non-
recycle modes were also calculated with pure us feed. In addition to
the 30-year histories, equilibrium cycle calculations were also performed.
In all these cases, the value of U233 was based on that value which gave
the same fuel cycle cost, independent of whether the reactor operator
sold bred fuel from the initial Th-u235 cycle or recycled the bred fuel
in subsequent cycles. The overall results could be well correlated by
assigning a value to pure u233 of about $18/8, and apply!ng a penalty
for any u236 present at the rate of about $13/8 U236.
-
-
-
-
-
.
-
-
-
*
-
-
-
- *
- - -
..
.
-
t24.
Since the above value 18 based on equal fuel cycle costs, using
U233 at $18/8 as makeup feed in the RIGR would give the same costs as
employing us as makeup fuel at $12/8; however, the nuclear performance
would be improved, with the conversion ratio increasing by about 0.05.
Additionally, the fuel cycle cost would decrease slightly (about 0.1
mai11/kwhr(e) 11 u233 were available at $14/8 as makeup fuel. Relative to
MSBR parformance, sale of bred U233 at $13/& Instead of $14/8 would decreass
· fuel cycle costs by le88 than 0.03 mill. kwhr(e). These values indicate .
that the nuclear performance of these systems can have a significant in-
fluence on reactor fuel cycle conditions only 11 other items of the fuel
cycle cost are small.
Fuel Utilization
Based on the 1962 and also the 1964 AEC projections for nuclear power
growth, large blocks of power will be required, such that by the year 2000
the nuclear power level will correspond to 730,000 MW(e); also, the rate
of growth at that time will correspond to the installation of at least
sixty-81x 1000-Mw(e) plants per year. If light-water reactors are used
with plutonium recycle, all the low-cost uranium supplies will have been
consumed by the year 2000, with about $12 billion expended alone for mining
natural uraniur (assumes 800,000 short tons of U20s available as low-cost
ore). To this must be added separative work requirements and the fabri-
cation and fuel handling costs, such that at least $30 billion total 18
Involved. Thus, it is evident that many billions of dollars are involved
relative to the utilization of nuclear fuel, and that economic reactors
having improved fuel utilization characteristics can be Justified on the
b&818 of savings associated with decreased fuel requirements. In porticular,
there is a large economic Incentive for development of breeders having high
specific power, low fissile-fuel doubling time, and low power costs.
Relative to fuel comitments, a most important factor is the fuel
utilization of reactors built prior to the use of breeder systems. Specifi-
and specific powers of 4 kg Pissile per megawatt electrical were utilized,
the fuel burnup.alone would be greater than 8 x: 106 kg of u235 over a
12
f
--
20-year cycle based on the nuclear power level projected for the year 2000
(to this must be added the fuel luventory of 2.8 x 10 kg of u235). How-
ever, only about 3 million kilograms of U235 would be recovered from the
800,000 short tons of natural. U308 estimated to be obtainable for $5 to.
$10/10. About 3 million kilograms of U235 would be recovered from U308
costing $10 to $30/10, and the remainder would have to be obtained from $30
to $50/10 U308.
The above indicates the need for early development of reactor systems
that have good fuel utilization characteristics. Indeed, using information
aid arguments similar to the above, the technological development of
advanced converters can be economically justified prior to breeder operation.
However, breeder operation appears required, and the following sections
consider how thorium reactors can satisfy fuel utilization needs.
In evaluating fuel utilization, the term has to be associated with a
specified goal. Here, it 18 assumed that good fuel utilization 18 associ-
ated with total f1881le fuel needs which require mining no more than cne
million tons of U30s. This is equivalent to requiring no more than 3.5
million kiingrams of U235 in the form of highly enriched uranium. A finite
value for mined fissile needs implies the need for breeder reactors which
eventually produce as much fissile material as is required to sustain the
economy.
In order to determine mined fuel requirements, a nuclear power growth
curve is needed, along with the fuel utilization characteristics of reactors
under consideration. Reactors which will be considered here and their
assumed characteristics are given in Table 5. The HIGR 18 the one con-
sidered in the advanced converter evaluation, with use of Bed to help
improve the breeding ratio; the MSBR 18 based on results obtained from a
recent design study;' the molten-salt converter-breeder reactor is the MSBR
initially operated with U235 fueling; the performance of fest breeder reactors
(Pu fueled), FBR, and fast converter reactors (U235 fueled), FCR, are based
on estimates for these systems. The nuclear power growth curve is based on
AEC estimates and 18 shown in Fig. 1; also given in Fig. 1 is a quadratic
fit to the curve as well as a linear fit. For all three curves, maximum
fissile fuel needs will be about the same if such needs occur between about
2000 and 2015.
-
-
Table 5. Fuel Utilization Characteristics of
Reactors Considered
8
(coubling
Reactor
CR
(Conversion
Ratio)
(Fuel Specific
Inventory)
[kg 11881le/Mwley
Time)
(yr)
0.9
0.8
HUGR
MSBR
Molten-salt converter-
breeder reactor
Fast breeder reactor
Fast converter-breeder reactor
0.8
1.0
For orientation, Fig. 2 shows fuel inventory and burnup requirements
as a function of time for two specific fissile loventory values, and also
for a change in breeding ratio (ABR) of 0.2 considering two thermal effici-
encies. If the ABR 18 an increase with the initial BR ? 1, the shaded
area represents a fuel credit; 1f the ABR 18 a decrease with the initial
BR - 1, the shaded area represents a fissile friel requirement. Note the
importance of specific inventory on fuel requirements; if the specific
inventory is low, the breeding ratio does not have to be much above unity
in order to give low net mined fuel requirements. At the same time, if
the breeding ratio is less than unity, burnup requirements will become
significant in the long run. Thus, based on an average conversion ratio
of 0.80 from now until the year 2020 and 25% thermal efficiency, over $100
bill.lon would have been associated with mining natural uranium for fuel
burnup needs. The same amount would he required to inventory break-even
breeders at 2020 having a specific inventory of 4 kg/Mw(e).
First, let us consider fissile fuel needs on the basis that there are
adequate amounts of plutonium or U233 available to inventory breeder reactors.
Under such circumstances, the maximum "mined" fissile fuel needs are given
byº the expression 1.56 x 10°SD2 (kg liesile), where D 18 fuel doubling time
in years and $ 18 specific inventory in kg f188ile/Mw(e); this corresponds
ORNL.DWG 11876

2000
- AEC ESTIMATE
---P: Q, (t-1989)
- -P=02(t- 1975)2
Qq = 66 X 103 Mwle)
Q2 = 1.17 x 103 MW(e)
lyr) 2
INSTALLED CAPACITY (103 Mwle))
OL
1970
1990
2010
2030
YEAR
NUCLEAR POWER GROWTH ESTIMATE
FIG. 1
ORNL-DWG 7,8011951

P: 1.17 x 103 17-1975)2 Mwle)
14 kg fissile
| ton U308)
S, .4 kg FISSILE
Mwle)
FISSILE FUEL REQUIREMENTS (106 kg fissile)
1.0
ABR = 0.2 -
- EFF. = 25%
- EFF. : 40%
NATURAL U REQUIREMENTS (100 tons U308
0.5
kg FISSILE
Mwle)
1970
1980
1990
2000
YEAR
2010
2020
2030
INVENTORY AND BURNUP FUEL REQUIREMENTS
FIG. 2
to about 0.5 million kg "mined" 11882le for either MSBR's or FBR's. Such
needs do not cause any resource problem, and so major mining needs appear
associated with startup of breeder systems.
HIER systems would normally recycle all bred U233, 80 such reactors
would not furnish fuel for breeders. If such systems were to be installed
for all nuclear capacity until 1990, and all plants bad a 30-year life, the
comitted mined fuel requirements for inventory and burnup would be about
1.4 x 106 kg U235. Corresponding requirements for installation oX HIGR's
up to 1995 or 2000 would be about 2.5 x 106 or 3.9 x 206 kg U235, respectively.
Thus, from the viewpoint of good fuel utilization, HTGR's need to decrease
their specific inventory and/or increase their conversion ratio, or not be
Since u233 would not be available for MSBR startup, such systems would
be initiated as converter reactors. Based on time dependent studies,' the
average conversion ratio during converter operation would be about 0.8, and
the time required to convert to a breeder cycle would be less than 3 years.
Under these circumstances and assuming a linear power growth rate, the
maximum Pissile fuel needs would be given by:
(D + 27) + B.u.
(1)
where a is the linear power growth rate, Mw(e)/yr;
s, is the specific fuel inventory required for converter
operation, kg fissile/Mwle);
D is the fuel doubling time for breeder operation, years;
q is the time required for a reactor to convert from
converter to breeder operation, years;
b.u. is the net Pissile burnup during converter operation, kg.
Alternatively, a lower limit for maximum mined lissile fuel needs is given
by:
.77D + 27) + B.u.
Also, the net fuel burnup requirements are nearly proportional to:
(1 - CR)aDT
17
Using expressions (1) and (3), the maximum mined f188ile needs for
MSBR's are estimated to be 0.86 million kg U255 for converter Inventory
plus 0.61 million kg U23s for converter burnup requirements, or a total
mined f18eile need of about 1.5 million kg uzzo. Such a need corresponds
to excellent fuel conservation, recuiring less than 500,000 tons of
natural U30s. Also, power costs during converter operation (3 years)
would be about 3.0 mills/kwbr(e), and would be about 2.7 mil18/kwbr(e)
during breeder operation? (27 years), based on privately owned plants
and a 30-year life.
Consider now other alteruatives. Hanford bas made extensive studies
of mined liisile fuel needs on the basis that fast breeders are fueled
with plutonium. Either HWR-U or PWR-U thermal reactors were considered
to initially supply plutonlum fuel for the fast breeders. Assuming that
fast breeders with a doubling time of 20 years were introduced in 1980,
and that fast breeders with a doubling time of 7 years were introduced
in 1990, Neef and Jones obtained maximum mined f1881le needs in excess
of 2 million tons U20s (equivalent to about 7 million kg U235, on same
basis as above). In terms of the criterion used here, such requirements
cannot be termed good fuel utilization. However, fast renctors can be
started in analogous fashion to that considered for molten-salt systems,
fueling with U205 fuel and operating as converters prior to becoming
breeders. Under such circumstances, the change from converters to breeders
could take place in about 5 years. Using the FCR and FBR characteristics
given in Table 5 and expression (2) for determining mined fuel needs, the
minimum mined fissile fuel needs are 2.9 million kg fibrile. Thus, it
appears there is better fuel utilization if fast breeder reactors are
used initially as converters to generate breeder fuel, rather than require
that fast breeders be fueled with plutonium.
The mined fuel requirement under the latter circumstance comes within
the term "good fuel utilization" as considered here; even so, the fuel
requirement 18 twice as large as that for MSPR-type systems. Also, the
relatively high fissile inventory and reduced breeding ratio of 1235-fueled
fast reactors would be detrimental economically, such that fuel. cycle costs
would be at least 0.5 mili/kwar(e) higher than the corresponding plutonium-
fueled breeder. Thermal thorium breeders such as the MSBR, however, pay
almost no inventory penalty in starting on 1235, and the burnup penalty
18 only about 0.3 mill/kwbr(e). Also, the FBR converter would operate
wider "peiralty" conditions for 5 years, while the analogous period for
MSBR converters would be 3 years. Thus, the best fuel utilization appears
potentially attainable with MSBR application, and the associated power
costs appear low.
In general, other thorium reactors which attain the specific fuel
Inventory and doubling time of the MSBR would give the same fuel utilization
results. However, they need to do so at low fuel recycle costs. Based
on present information, HIGR and HWR-Th systems need development of improved
fuel fabrication and processing techniques which will give low fuel recycle
costs at high fuel processing rates in order to achieve the performance
characteristics of MSBR systems.
An additional consideration in fuel utilization studies is the separa- ,
tive work required to meet fissile fuel needs. Relative to use of MSBR
systems, the maximum required rate for highly enriched uranium would be
about 66,000 kg U265 per year for inventory requirements plus 30,000 kg
v235 per year for burnup requirements, or a total of 96,000 kg 1235 per
year. This corresponds to maximum separative work needs of about 23 million
kg SW/yr, a requirement which can readily be met by present capacity
(estimated to be about 21 miliion kg SW/yr) plus a small amount of stock-
piling if needed.

19
References
1. M. W. Rosenthal, 8. F. Bauman, L. L. Bennett, R. 8. Carlsmith,
and D. R. Vondy, The Technical and Economic Characteristics of
Thorium Reactors, ORNL-IM-1145, June 3, 1965.
2. M. W. Rosenthal et al., A Comparative Evaluation of Advanced
Converters, ORNL-3686, January 1965.
-3. R. S. Carlsmith, C. M. Podeweltz, and W. E. Thomas, Fuel-Cycle Cost"
Comparisons for High-Temperature Gas-Cooled Reactor Frels,
ORNL-IM-1112, April 14, 1965.
4. L. 1. Bennett, A Study of the Fuel Value of U233, ORNL-TM-1499
(to be 1bsued),
5. Paul R. Kasten, E. S. Bettis et al., summary of Molten-Salt Breeder
Reactor Design Studies, ORNL-IM-1467, March 24, 1966.
6. Paul R. Kauten, "Nuclear Fuel Utilization and Economic Incentives,"
Trans. Am, Nucl. Soc., 8(2) (Nov. 1965).
7. K. J. Yost (ORNL), private comunication, March 3, 1965.
8. W. I. Neef and E. D. Jones, Jr., "Conservation Economics and Reactor
Technology," Trans. Am. Nucl. Soc., 8(2): 576 (Nov. 1965).
9. L. Reichle, Nucleonics 22, p. 19 (Sept. 1964).

LW
V wLWN
.
..
.
-.
LE
END
*
-
DATE FILMED
7 / 29 /66





-


-
-
VA
IN
...
.
.