i-.-'chq—‘: . . L)b\l.'&)d

OAK RIDGE NATIONAL LABORATORY MA@TEH QQ ?Y

Operated by '
UNION CARBIDE NUCLEAR COMPANY
Division of Union Carbide Corporation 0 R N L
@ CENTRAL FILES NUMBER

 

Ock Ridge, Tennessee

 

 

DATE: October 17, 1958 copY NO. S

sUBJECT: Survey of Iow Enrichment External Transmittal

: Molten-S8alt Reactors : ‘ Authorized

TO: Tisted Distrivution Distribution Timited to
_ Recipients Indicated

FROM: H. ¢. MacPherson

A rough survey of the nuclear characteristics of graphite-moderated
molten-salt reactors utilizing an initial complement of low enrichment
uranium fuel hag heen made. Reactors can be constructed with initial
enrichments as low as 1.25% U-235; inltial conversion ratios of as high
as 0.8 can be obtained with enrichment of less than 2%. Highly enriched
uranium would be added as make-up fuel, and such reactors could probably
be operated for burnups as high as 60,000 MWd/ton before buildup of fis-
sion products would make replacement of the fuel desirable. A typlcal
circulating fuel reactor of this class might contain an initial inventory
of 3600 tons of 1.8% enriched uranium, coperated at 640 Mw (thermal), and
generate a net of 260 Mv (electrical). The total fuel cycle cost would
be approximately 1.3 mills/kwhr, of which 1.0 mill is burnup of enriched U-235.

NOTICE

This document contains information of a preliminary
nature ond was prapared primarily for internal use
at the Oak Ridge National Laboratory. 1t is subject
to ravision or correction and therefore does not
represent a final report,

The informatiog {n not v be 5. ';;.?;.et;:aui = RELEASE APPBUVEH
reprinied or ofnerwise ?_;_-jx;c 1 oubhe diseog a-‘fl"?«“
withoot the approval of e . SN st b BY PATE I BflANGH

Legal asd Laxwt&en Lasle ol Jlemr? weed

 
Errate for CF 58-10-60

Please make 'l:hé following changes 1n your copy of CF 58-10-60:
page 1 (Cover Sheet) Make change in line 10 of abstract
| . From: 3600 toms of 1.8%
To: 36 tons of 1.8%
Page 2 Mske change in next to last line

From: 8.5k x 1020

To: 8.54 x 1022
Page 3 Make change in line 18

From: Jl

To:

St

G,%, ST
D
SURVEY OF I0W ENRICHMENT MOLTEN-SALT REACTORS
To survey the field of low enrichment graphite-moderated reactors

a number of calculations have been made with the four-factor formula k_,= nepf.

The following fuel salt was considered:

mole %,

UF, 20
11 (F .70
BeF,, 10

This salt has a melting point of about_900?F. It is probably more corrosive
than one mole % fuel, but 1s probably satisfactory for use with INOR-8. The
atomlc concentrations were as follows at 600 - 65OOF, per L. A. Mann:

Ii - 175.7 x 10%° atoms/cc

Be ~ 25,12 x 10 atoms/cc

U ~ 50.22 x 1020 atoms/cc
P - 426.8 x 10°° atoms/cc

t

The slowing down power of the salt is 0.0228 cm"l, composed of contribution

as follows: =

Fluorine 0149
I L0051
Be 0028

The concentration of UF) is 2.62 g/ce; that of uranium is 1.98 g/cc. Graphite
was assumed to be of density 1.7, with 8.5k x 1070 atoms/cc, and a ‘slowing down

-1
power of 0.06%1 ecm ~. The graphite was assumed to have a o, = 0.00L45 barns.
-5

In the four-factor formula, n is taken for convenience to be that
for U-235, while the thermal utilization factor f is defined as the propor-
tion of thermal absorption in U-235. Thus, using cross sections derived
from PFig. 2,5,7page 2.20, of ORNL~2500, Part 2, n for U-235 is taken.as
2,47 X_g%g or n = 2.00. Effective thermal fisslon cross sections and absorp-
tion cross sections are assumed to be 528 barns and 650 barns, respectively,
corresponding to neutron temperatures of GOOOCo

The fast fission factor was assumed to be 1.02, since the values
calculated for graphite lattices vary from 1.02 to 1.04. This aéSumption
is sufficiently accurate for survey purposes.

For the calculation of the resonance escape probability, the reson-

ance integral was calculated from,

0.k2
8 S
0. = 3.8 N + 24,7 ¥
o
Zs
In this formula, T is the scattering cross section in barns per uranium
0
atom within the fuel channel, and 5 is the surface area of the fuel channel

M
per gram of U-238 in the channel. The first term is the same as the reson-

ance integral for an infinite medium of the fuel salt composition. This
formula, involving the first power of %-instead of q-% , was used because
it is more logically extrapolated to the case of dilution of uranium with

fuel salt. For uranium metal this reduces to the familiar dr =9.25 x 24,7 % .

In the fuel considered here, “s = k3.5 , and o_ = 18.5 + 2b.7 2 .

NO
e

A series of reactors having tubular fuel channels on a square
array 1s considered. The spacing of thése channels was arbitrarily taken
as 8§ in. center to center, and the calculations are made for various volume
fractions of fuel in thé graphite from 0.05 to 0.25. The following table
gives the fuel channel diameter, the value of % » the value of 0. and the
value of the resonance escape probability p for the different volume frac-
tions. p 18 calculated from p = e-A where A is glven by,

Nu . F

A=
(1 - F)(eng), + F(ES) o0y

 

where N 1is the U-238 atom density in the fuel channel and F is the volume

fraction of fuel in the core. Numerically,

 

 

 

s o 90.22x 1020 o. Fx 10”3“
0.0631 (-1-F) + 0.0228 F
Table T
F D 8/M °r D e
.05 2.08 in. . 382 27.95 barns 0.891 1.82
075 2,5k .313% 26.25 0.848 1.7%
.10 2.94 271 25,20 | 0.807 1.65
«15 3,60 221 0%.97 0.733 1.495
.20 4.15 .192 23,25 0.653 1.%32
.25 h.65 171 22.73 0.58Lk 1.191
The thermal utilization factor is given by,
FeN o
f = 281,18.
/\ i
FeN o +FN o Loy zasalt +(L-mzfy
*

-5~
where e is the enrichment, g;u is the effective absorption cross section for
U-235, and 7 1s the thermal flux disadvantage factor. For this survey, y is
assuned to be 2.0, which seems a reasonable value based on ORNL-2500. This
simplifying assumption was made to avoid calculating the flux distribution,
and probably 1s the roughest approximation made in this survey. Numerically

the above equation reduces to,

5.22 F e

P =322 F e +0.0037 F + 0.0085 F 7 0.000788 (T =F)

An easy computationsl form is,

 

0.00077
% =1 4 0.0157 + %
5.22 e

To calculate the enrichment necessary to achieve a given k__, the following

transformations are made,

 

ko = nept
1 . nép
F ok,
0.00077
3,00 CEEE - 1)
oo

The conversion factor is calculated as follows,

R - {1 - p)+ p x{proportion of thermal absorptions in U-238)
e p x(proportion of thermal absorptions in U-235)

From k_, , 8% 1s obtained from B2 = Feo™ 1

e ————

o
and M? is taken as,

M =T graphite + 12 hite ¥ (proportion of thermal absorptions
1-F grap cceurring in graphite)

is an approximate one, but not too

. 2 2 . o
is taken as 324 em™ and Igraphite is 2950 em .

The correction for’Téraphite

important numerically. 7Jgraphite

From B2 the dimensioms of a mipimum volume cylindrical core were cal-
culated using a reflector savings of 2 ft. This is less than used in ORNL-2500,
but may still be optimistic because some fuel must be used in the reflector to
cool it.

Table IT gives the resulis of calculations for six values of the
volume fraction of fuel in the core and for two values of k.. The volume of
fuel in the core should be evaluated in terms of the external volume for g
cireculating fuel reactor, which is about 0.56 cu ft per thermal megawatt, or
360 cu ft for a 260 Mw (electrical) plant. It is evident that one must pay
for conversion ratios above 0.8 with enrichments of over 2%.

The selection of an economically optimum reactor of this type
requires a knowledge of the method of chemlcal reprocessing and its cost,
and a way of calculating the effect of poisoning by buildup of fission pro-
ducts. The latter problem was looked at briefly, using as a basis the calcu-
lations of Blomeke and Todd (ORNL-2127), and assuming a buildup of Pu sub
infinity as defined in the Brookhaven Ceneva Paper 461. For the very long
exposures 1t 1s brobable that the one~group cglculations can not give a
good answer becausé of the large buildup of absorbing Pu isotopes. However,

it might be possible to operate a reactor such as Case 8 of Table II for as
2

-7~
long as 30 years at 640 Mw (thermal) without a need for the U-235 inventory

to increase by more than a factor of two, and with a breéding ratio averaging
greater than 50%, without any reprocessing.

For purposes of estimating the fuel cycle cost, a life of 10 years
without reprocessing was assumed. For this period, the U-235 concentration
would probably not have to be increased over its initial value, and the breed-
ing ratio should average at least 60%.

Applying the formula of ORNL-2500, but using the thermal cycle of the
reference design molten-salt reactor (Clb"~5_8-~5-5)‘9 the chemical reprocessifig
charges and fuel inventory charges are 0.034 mill/kwhr and 0.14% mill/kvwhr,
respectively. Ten-year depreciation and capital charges on the base salt
amount to 0.093 mill/kwhrn Burnup of U-235 would be wrl1l.0 mill/kWhr at a
conversion ratic of 0.6. Thus, total fuel eycle costs would be approximately
1.3 mills/kvhr.

This reactor was based on Case 8 of Table I, using a total fuel

volume of 600 ft5

and an inventory of 36 tons of uranium of 1.8% enrichment.
The ten-year reprocessing cycle represents a fuel life of approximately
60,000 Mwd/ton.

More accurate calculations are needed to confirm the above conelusions.
_8-

 

Table IT
Vol fraction Percent Vol of fuel Uranium Critical Initial
of fuel Enrichment  in core in core mass of conversion
in core of uranium cu £t kg U-235 Ratio -
Case F k,a e Vf Mu M255 -Rc
1 Q.05 1.05 1.30 395 22,100 298 546
2 1.10 1.45 143 8,000 116 492
3 0.075 1.05 1.25 kot 23,900 298 635
L ' 1.10 1.%9 167 9,350 130 600
5 0.10 1.05 1.275 bhs5 ok, 900 318 « 70T
6 1.10 1.46 179 10,000 16 668
7 0.15 1.05 1.525 LY 26,600 405 796
8 1.10 1.80 197 11,000 198 . 780
9 0.20 1.05 2.24 520 29,100 65é 865
10 1.10 . 2.88 206 11,550 332 865
11 0.25 1.05 k.26 575 32,200 1400 .900

12 1.10 7.05 240 13,4%%0 952 .865
l’

e

O O=~] WA

11.

12,
13.
1L,
15.
16.
17,
18.
19.
20.
1.
0D,
0%,
ok,
5,
26.

C.
E.

F.

W.
E.

D.
W.
G

D.

W.
Ha

H.

» G.

Jo
Se

» P.

F.

@ L-

FC
Je.

oEs

a.
H.
I.

-A-

A.
X.
L.

. P.

R.
We
H.
W

A.
Je

Alexander
Barton
Bettis
Blakely
Blankenship
Boch
Boudreau
Breeding
Browning
Campbell
Carr
Cathers
Charpie
Douglas
Ergen
Falkenberry
Fraas
Grimes
Hoffman
Jordan
Keilholtz

. Kertessz
- w‘
. E.

Kinyon

Iackey

Tane

larkin, AEC, ORO

_9_

Distribution

" 28.

29,
%0,
%1,
32,
35.

25
26.

580'

59
Lo,
hl.
42,
L3,
L,
L5,

17,

L9,
50.
51.
52.

MacPherson
MacPherscn

Manly

Mann

Mann

McDonald

Metz

Milford

Moesel, AEC, Washingbton.
Nessle '
Osborn

Roberts

Savage

A. W. Savolainen

M. J. Skinner

E. Storto

J. A. Swartout

A. Taboada

R. E: Thoma

D. B. Trauger

F. C. Vonderlage

G. M. Watson

A. M. Weinberg

G. D. Whitman

J. Zasler

Iaborstory Records, R.C.

H. G.
R. E.
W. D.
E. R.
L. A
W. B.
Ho J.
R. P.
F. C.
G. Ja
W. R.
J. To
H. We
L | “p 37

Errata for CF 58-10-60

Please make the following changes in your copy of CF 58-10-60:
Page 1 {Cover Sheet) Make change in line 10 of abstract
. From: 3600 tons of 1.8%
To: 36 tons of 1.8%

Page 2 Make change in next to last line

From: 8.54 x‘lOEO
To: 8.54 x 1022
Pé.‘ge 3 Make change in 1ine 18

From:

To: \}

z:m\ i

AT