EIR-249.txt

EIR-Bericht Nr. 249

EIR-Bericht Nr. 249

Eidg. Institut fir Reaktorforschung Wirenlingen
Schweiz

A Molten Salt Fast Thermal
Reactor System with no Waste

M. Taube

il

Wurenlingen, Januar 1974

EIR-Bericht Nr. 249

A Molten Salt Fast Thermal

Reactor System with no Waste

M. Taube

January 1974
Abstract

A Power system of 1 GW(e) containing a fast breeder reactor
(2GW(t)) with a molten chloride fuel and a thermal burner reac-
tor (~ 0,5 GW(t)) with molten fluoride fuel resulting in a

total breeding system with a doubling time of 30 years 1s des-

cribed.

The fast breeder works on a U-238/Pu-239 cycle, the thermal
burner on Pu-23%9. The thermal burner ‘'incinerates' (by neutron
transformation) most of the long lived fission products. Kr=-85,
Sr-90, Te-99, Cs-135 and other. The actinides (Np, Am, Cm)
might be recovered from the molten salt with a high efficiency
and then burnt up in the fast reactor. The entire system pro-
duces no long lived radioactive wastes with the exception of

T and partially Cs-137.

The fast reactor has an inherent stability against power excur-
sions, and loss of coolant accidents because of the two or more
independant cooling circuits and in which the fertile material

(molten uranium chloride) plays the role of cooclant.

A very high efficiency aluminium-chloride secondary coolant and
turbine working agent is proposed. Together with a district

heating system practically no thermal pollution occurs.

A very low I-131 and Xe-133 and low Cs=-137 concentration, and
very low CsCl volability in the steady state core due to con-
tinuous gas extraction guarantees the high safety level of this

system.
I. Why a new reactor system?

The present generation of power reactors both thermal and fast
both liquid—cooied (water or sodium) and gas-cooled, use solid
fuel. The single reactor type with positive experience of liquid
fuel is the Molten Salt Breeder Reactor (thermal reactor with
molten fluoride and with breeding ratio ~1.05) (Rosenthal, 1972).
Some other molten salt fast reactors have also been dlscussed

in the literature over several years (Taube, 1961, 1967,

Nelson 1967).

The liquid fuel molten salt reactors have some specific charac-
teristics which make them more attractive for future use, espe-
cially in form of a coupled system of two molten salt power

reactors as 1s proposed here

1) a fast breeder reactor with molten chloride fuel

2) a thermal burner with molten fluoride fuel.

This system appears to have the following special features which
results in a reactor unit much more fitted to the future pattern
of energy production than are the present day solid fuel reac-

tors:
1) high inherent stability

2) much smaller environmental hazard

%) elimination of long lived [ission product wastes and other

radioactive wastes

4) better suited to district heating systems with a rela-

tively high thermal efficiency for electriclity generation.
A

The "fast breeder - thermal burner" system proposed here has

the following specific characteristics

1) inherent stability against power excursions which arise
out of the following chain of events: criticality increase~
power excursion - temperature increase - density decrease »
movement of part of the fuel out of the core » very strong

negative influence on the criticality.

2) 1in the fast breeder reactor the loss of coolant 1s at the
same time the loss of blanket which results in a very high
negative coefficient of reactivity in this "Loss of Coolant
Accident™. The reactor thus shuts down without an engineered

scram.

3) continuous purging of the liquid fuel Iin the core by means
of a hydrogen/helium stream for extraction of the vola-
tile fission products especially the most hazardous I-131
and Xe-133% and also I1-137, Xe-137, the volatile precursors
of Cs-137 which have a large effect on the environmental

impact in any reactor accident.

4y  continuous reprocessing of liquid fuel with extraction of
the long lived fission products (e.g. Kr-85, Sr-90, Zr-93
Tc-99, I-129, Cs-135) and the burning (neutron transfor-

mation) of them in a thermal reactor.

5) continuous reprocessing (with high yield) of the actinides

(Np, Am, Cm) and continuous burning 1in the fast core.

6) the possibility of having a combined breeding ratioc for
both the fast and thermal reactors coupled together,
greater than 1, achieving a doubling time of approx.

30 years.
7) with the exception of tritium, virtually no nett rejec-
tion of radiocactive wastes to the environment, 1f Cs-137
has been stored practically all the waste 1is transformed

in the form of stable or semi-stable nuclides.

8) some of the corrosion processes of molybdenum (structural
material) occuring in the core and blanket may be control-

led by the continuous gas purging system.

9) the high thermal efficiency for electrical energy produc-
tion of the total power system (55 % for only electrical,
or 40 % for combined electrical and district heating
system) results from the use of a chemically dissociating
medium as secondary coolant and turbine working agent
(AlClB).

10) the system would seem to be suitable for an underground and

fully automatic power plant of high safety.

11) this type of reactor gives, as do other molten salt systems,
a complete independance from foreign supplies (no fuel-
element manufacture, no external reprocessing plant) and
has the ability to 'burn-rocks' - i.e. low grade uranium
"ores" from granites.

(see Fig. 1)
I-131, Xe-135%

Heat energy

. Granite Cores W fi‘\\ Granite waste
~100 tonfday preparatioij - 441-7
~100 ton/day
U,
VTh
Gaseous
/f reprocessing
Y
liquid fuel N U233 0,%kg/d
process and —— > (238 0,3kg/d
~ blanket fuel dP———g—
preparation A
Np
s ¢ —1 N\ o
1000 MW(e)
Fast Electricity
very stable \ breeder V
i%g;“St ~2000 MW (t) no thermal
polution
in steady Stgfifi// Heat excnan A
very small ] ger < P
amount of [~ coolant/A1{] 1100 MW(th)
\

Cs-137

i

Thermal
burrier

~500 AW (L))

F1ss

) . Producks
reprocessing

(low temperature)

)
Cs-157 angw

N

T storageJ

stable Nuclides
(with some amounts
of long lived)
II. The fast breeder - thermal burner coupled system

The total power system of approx 1 GW(e) contains:

1) fast breeder reactor with molten plutonium-chlorides fuel
cooled in core by means of fertile material (uranium-238
chloride as fertile and coolant) The thermal power 1is

approx 2000 MW.

In the coupled fuel cycle the breeder with a breeding
ratio of ~ 1.4 form the basis of the breeding ratio of
the entire system of 1.1 giving a doubling time of approx
30 years. In the fast reactor are also some of the long

lived fission products being held under the neutron flux.

2) thermal burner reactor with molten plutonium fluoride
fuel, cooled externally (out of core). The thermal power
is approx 500 MW. In the coupled fuel cycle the burner
(with breeding ratio equal to zero; no fertile material)
is responsible for the burning of the longest lived
fission products coming from both reactors fast and thermal

that 1s from total power of 2500 MW(t).

3) an appropriate power and heat generating system with the

following circuits

reactor fuel - in core cooling agent (molten uranium Salt)

secondary coolant out of core (aluminium trichloride)

tertiary coolant (e.g. nitrogen dioxide)

hot water for district heating.

4y multi-stage multicomponent reprocessing system with the

following units.

- reprocessing of the fast core fuel, including prepara-

tion of fresh fuel
- reprocessing of the blanket material from the blanket

of the fast reactor

- preparation of uranium chloride from natural uranium

oxide

- reprocessing of the thermal core fuel
- reprocessing of the irradiated solid or molten long

lived fission products
- reprocessing of the in-core gas purge.

(see Fig. 2)

h:]

Since the system contains both breeder and burner reactors and
since a combined breeding ratio greater than 1 is required the
relationship of the power of the fast breeder and thermal burner

1s the highest importance.

The problem of the achievement of the breeding ratio for the
total system can be solved by the following simplified calcula-

tions concerning the ratio between the respective powers.
Nuclides

Fast/Thermal Reactors System
CHEMICAL l NUCLEAR THERMAL/ELECTRICAL
~3.0kg/d 0.3%kg/d
v 4
STABLE F.P.
Pu Long Fiss =
lived {Prod
rad oac%eplflO
Fessipg lHeat [€{Heat p——
 J % Y — % Ex.
Fuel >
> JCoo1lingy >
| [he rm{*< Zone
Puel THERAAL] BURNER i Yo v ]
e D00 MW (1) T ing urbin
r_,\__,\__;fiepro ) 1C1 WO &
Drainage ] < 1000MW
\fi -< - —— > (e)
Fast rf”’ r/”'
11
é—fl\—— Fuel Fuel @}
S TRepro[Z 1> *\\\\~
(__/‘\__q\__>
A K__J (:)
Y 4y
e
Heat Heat
bs—P—F: lankpt Fxchnd |kxch ot
8 i3
L L 1P %nankde i) fio, fwater
[~ ! = 1| ° leccwi
L____fh__n\__a - >
COOLINY) |
FAST HRELDLR
Gas 2000 MW (L)
\
prain Tank
Stable Tritium
Table 1

Calculation of the fast breeder/thermal burner ratio

Unit Fast breeder Thermal burner

Total power MW(t) X 1
Specific power gPu/MW(t)

ln core Pf = 1000 Pt = 20

in total system

including repro-| gPu/MW(t) Pf = 1100 Pt = 80

cessing
Plutonium burning gPu/MwWd(t)| F = 1.1 F = 1.1
Breeding ratio gPu/MwWd (t) Bf = 1.35 Bt = 0
Losses 1n the re-
processing system gPu/MWd(t)| V = 0.05 V. = 0.05
and uncertainties
Breeding gain gPu/Mwd (t)

G=F.(B-1-V) F.(1.35-1.0-0.05)=| F.(0-1.0-0.05)=

= + F, 0.30 = ~F, 1.05

Postulated
doubling

time: T2 years/days = 30 years = 10 days
leeded breeding gtotal _ yx p 4+ p 107

: f t .
galn = — = 0.1

I
T2 10

Effective (total breeding gain for fast/thermal system:

¢ °% - x-F 03 - F-1.05 = 0.
The fast/thermal power ratio: X 1.00 + 0.1 5 E

0.5

10

For a self-breeding system containing fast breeder and thermal

burner the following power distribution is needed (B = breeding

ratio)
ffast
- fast breeder reactor (B = 1.,35) b MW(t)
- thermal burner (Bther = 0) 1 MW(t)
- total coupled system (BSySt 1.1) 5 MW(t)

or for a total power of 2500 MW(t):
- fast breeder: 2000 MW(t)

- thermal burner: 500 MW(t)

A very rough estimation of the characteristics of both reactors

is given in Table 2.

More details about the fast reactor are given in Chapter 3

and the thermal reactor in Chapter 4.
Brief characteristics of the

11

Table

2

"Fast breeder/thermal burner"

system

Total electrical power

Total efficiency (arbitrary)

Total thermal power

Reactor, total power

Specific power (per unit
volume)

in core

in reactor with coo-
ling region but with-
out blanket

Core volume

Cooling region volume

Neutron flux

Core volume composition
Fuel-liguid

Moderator

Coolant

Fission products

Tubes

MW(t)

MW{(t)/1litre

=1 GW(e)
= 40 ¢
= 2.5 GW(t)

Fast Breeder

Thermal Burner

2000

8750
same as core

T X 1015*

38.6

none

55.5 (in this
case the fer-
tile material)

no special vo-
lume for F.P.

5.9

500

1.1

homogeneous fuel;
BeF2 as moderator
no coolant in the
core; external
cooling

84 in the core

Blanket

Role in fuel cycle

Breeding ratio

Fuel components

fissionable

diluent
moderator

Irradiation target
(fission products)

Literature

2

47.85
(UCl3 as fer-
tile“material)
Breeding of Pu

in U-23%8/Pu=-239
cycle

~ 1.35

Molten chloride

Pu-239 + other
actinides (Am, Cm

NaCl
Nno

Zr isotopes (?2)

(Taube, Ligou
1973)

no blanket

Burning of long
lived F.P.

0

Molten fluoride
Pu-23%9

Zr Fu

BeF2
Kr-85, Cs-135,
Sr-90, Tc-99.
I-129

For thermal bree-
der reactor with
molten fluorides
ffuel see

(Rosenthal, 1972)

* here for calculation purpose a neutron flux of 8 x 10

lated, however a flux of 10 n

15

1s postu-

_2_. .
cm S seens reallsable.,

13

I1f. The Fast Breeder Reactor

In this study a 2000 MW(t) fast breeder molten chlorides reactor
was used. (Fig. 3 and 4)

The detailed describtion of the reactor type was recently given
(Taube, Ligou, 1973). The most important characteristics are

summarised here: (Table 3 and 4)

1) the liquid fuel contains only plutonium chlorides diluted
by other metal chlorides. The 1n core breeding ratio is
realised by the presence of the uranium-2338 chloride
(diluted by other metal chlorides) which acts as the coo-
ling agent flowing in the tubes.

2) the fertile material (233 UClB/NaCl) is at the same time

the cooling material and the blanket material.

A very important and desirable feature of this fast reactor can
be seen when tne "maximal design accident'" 1s discussed. Thils

appears to be the Loss of Cooling Accident (LOCA) (Fig. 5)
"he LOCA in fast reactor arises from two events.

1) the increase of criticality due to each coolant (even
helium) being a neutron absorber and the loss of coolant
results in an increace of neutron flux in the core (a
trivial polint 1s that 1In a thermal water cooled reactor

"thne Loss of Coolant" 1s also a "lLoss of Moderator'" whilch
reduces the criticality); thus in a (ast reactor the LOCA

must be accompanled by an engineered shut-down device.
14

. 3 CHLOROPHIL / EIR

2 GW(E)
>
Sl
fi%ooc 790° AN | | AR
e
NV, A
qulOm S;Z

)W WS OVAV )

3 | Coreo
/A/ /{/* ‘ ~9g90 " C | fuel

TN,
KRR

ninm

2.5b

15

EIR

CHLOROPHIL

4

Fi

AlCl
5

Blanket Material

Cooling Agent

3 independent cooling circuits

(including 3 blanket regions)

Core (not divided) with 3 pumps

Heat Exchanger

SO0

4,26 m
16

Table 3

Molten Chloride Fast Breeder Reactor (MCFBR)

"CHLOROPHIL"

Electrical power, approx
Thermal power, total/in core
Core volume

Specific power

Core geometry

Fuel: liquid PuClB/NaCl
Liguidus/boiling point

Fuel mean temperature

Fuel volume fractlion 1in the core
Power form-factors radial/axial
Fast flux, mean across core

Fuel density at 984 °c

Heat capacity "

Viscosity "

Thermal conductivity, at 750 °c
Fuel salt 1n core

Total plutonium in core/
in system

Plutonium in salt
Mean plutonium specific power

Mean plutonium specific power
in entire system

kg
welght %
MW(t) ke

1

~ 800
2050/1940
8.75

220

height 2.0/
2.36 diameter

16/18
.685/~1500 (approx)
G814
0.386
0.60/0.78
' 1015
2344
0.95
0.0217
0.007
7900

2900/3150

36 .4
0.07

0.62

17

Table 3

Coolant 1liguid U-238 ClB/NaCl Mol % 65/35
Liquidus / Boiling point °¢c ~710/~1700
Coolant temperature inlet/outlet °c 750/793
Coolant volume fraction in the 0.555
core )
Coolant density kg m™° 4010
Coolant salt in core/ in \ .
blanket kg 19'500/165'000
Thermohydraulics
Fuel (shell side, pumped), -1

. ms 2
velocity
Coolant, velocity msml 9
Number of coolant tubes 23'000
Tubes inner/outer diameter cm 1.20/1.26
Tubes pitch cm 1.3%8
?reedlng ratio, internal/ 0.716/1.386
votal
Doubling time, load factor
1.0/0.8 years 8.5/10.5

18

Table 4 Fast breeder neutron balance
Region Atoms Absorption Leakage | Production
e -1021 % % yA
- (n,y) 22.51
U-238 3.5629 25°5O(n,f) 2.99 8.2%
- (n,y) 5.58
Pu-239 0.66796 3M.56(n,f) 28.98 85.55
_ (n,vy) 2.24
Pu-240 0.16699 3.78(n,f) 1.5y h.72
Na 6.3017 0.26 -.-
[
0% in fuel 1.10
o 1 19.495 5 in c00l.2.06
e 5.978 1.30 -.-
Mo 0.7386 2.04 -.-
¥.P 0.0679 0.50 -, -
Total -
Corpe 71.10 27.40 98.50
- (n,y) 23.15
U-23%8 6,42 23'7O(n,f) 0.55 1.50
=
@ Na 3,457 0.08 - -
et
E c1 20.72 .00 .-
2a)
Total 26.00 2.9 1.50

Blanket

19

fi NN NN NN NN
NN N NN N NN
e —— fi.um
o
NN N\ \ \ N N (s}
<~ N\ N\ Wi N /// N Z
LN S N S N =
N O
S // < HM
N N o
N\ N TS
AN N < F o<
N N 2 e O R
AN N o AD D
// // | [ |
N
23
B = a0 m
AN
// N
= N \ =
N N
g N
= // NN
© N / =
Q AN / // =
< N AN —
AN N N / / =
= A
= A ) o\
@ // 52 ~ N\ N\ 2
S SEIEIRIREIICHEREKL N =
= h 0O R RRRREIELLRRL
N CRRERIREGRENRZX
S \ SRR o N
N\ ~ L N N
LRI
pm N b 3 WN@"OMOMOWO % hflflo &
N\ X X 0.,<c T e e RIS ISR KKK, J
) N nnouw«nawuu EECSSS TSNS EORRARREERELRL N
3 ~ O\ ROERI] N N
-] AN T~ N <
WO ///x ///////////(
" N \
NONON NN //
: N SN NN
A N
. NN NN N
I, N
NN\ NN

PRK

S S S S s

S s s s s

7

S S s s S

Vv
/

///
s S S s

e
N

TSI

N
SN

S N NI
LK RZLLXLX T

AY A N N
ERRZPX ™

NN N NN

,
IOV
\
NGO AN )
N o A
PSRN0 00w | N N
AN

404"»0n0u0¢)(’0b‘h0»0n0NOMOMON«
~ \ N\ \ \ S N q
RERLIR IR

L& N N N

s
///// 7

),

-

N
AN
N ////

OO

COOLANT
20

2) the radiocactive decay heat of fission products 1s approx
7 % of the normal steady state power level and is high
enough to melt the clad and in some cases the fuel in a
few tens of seconds from the LOCA, especially in fast

reactors with a high volumetric specific power.

The "Emergency Core Cooling Systems" are still under critical

review.

Here in this type of fast reactor, both "negative" effects of
LOCA could be virtually eliminated by means of the following

design features (not shut-down devices):

1) the cooling system is coupled with the blanket, that 1is
the cooling agent plays the role of fertile blanket

material.

2) the cooling system including the blanket region is divided

into two or more fully independant circuits.

3) it is postulated here that a major accident will cause com-
plete destruction of all but one of these independant cir-

cuit systems and one of the fuel pumps 1is intact.

4}  the cooling system sub-divides into several if not all

parts of the cooling region.

During an accident the loss of coolant accident thus brings about
a corresponding loss of blanket in the reglon affected. The loss
of blanket gives a very strong negative coefficient of reacti-
vity which is clear from Table 4. The roughly estimated value is
approw (7 + 1%)/N where N is the number of independant cooling/
blanket circuits. In this reactor a failure in only one out of
four circuits results in a reactivity decrease of (7 % 1%)/4

which is equivalent to a full scram.
21

Thus the reactor stops without the use of any engineered devices.
It seems that this is the only type of fast reactor with inherent

negative criticality against the Loss of Coolant Accident.

The radiocactive decay heat still has to be removed, if not, the
vessel and cooling tubes may melt down. In this reactor the one
remaining intact cooling circuit is able to remove this heat,

because the molten fuel acts as the internal heat transport me-

dium and assists the heat removal to the one intact system.

IV, The thermal burner reacEgg

As is shown in chapter VI the most effective neutron transforma-
tion of a large part of the long lived fission products is possib-
le in a thermal neutron flux because the products of cross section
and flux (Oj . Gj) is often larger for thermal neutrons (g,=

thermal) than for fast neutrons (see Fig. 16).

This is one of the reasons for using a thermal burner for the

vransformatilion.

Some general remarks about the criteria for this thermal burner

are given below:

1) the thermal neutron flux must be as high as possible

but as will be show there 1s an upper limit. It appears

that a flux of 3 - 6 X 10lbn cmnzs_l is most suitable
(e.g. for Oy 5 barns; o-@ = (5 x 10-2M) X
(6 x 1015) = 2 X 10_8&;—1 which corresponds to an effec-

tive half 1life of 2.25 years).
22

2) the specific volumetric power should not be too high
because the thermal core volume and the neutron balance
has to be kept to a size to accommodate the appropri-
ate amounts of neutron absorbing fission products (e.g.
specific power 1-2 MW/1litre) (for 1 MW/l the core volume
is 500 1, for 2 MW/l the volume 1s ~250 1).

3) the specific power per unit volume must not be toc high
to keep the total power down. The thermal core 1s intended

as a burner only, that is with breeding ratio zero.

4) the core containing only fissionable nuclide, moderator
and appropriate fission products must give the best possible

neutron balance. The following two postulates are added.
5) no fertile material in the core

6) no coolant in the core except for the fuel itself which
is cooled by means of out of core heat exchangers. The
fuel 1s also in 1ligquid form to permit continuous repro-

cessing.

As will be seen from the following discussions the apropriate
specific volume power for this core is choosen to be as high as

possible: 2 MW/1.

(HFIR Reactor in Oak Ridge, USA; 100 MW(t): 100 litre core; 1 MW/1)
CM-2 Reactor in Mellekes, Soviet Union 50 MW(t); litre; MW/1
both with mean neutron fluxes of 6 x 1015 N cm—gs—l. It must be
stressed that a more conservative figure of 1 MW/1 and flux of

3 X 1015 n cm_gs_1

would also meet the reguirements but of course
with lower effectiveness (higher steady state concentration of
fission products). Fig. 6 and 7 give some detalls concerning the
thermal high flux burner. Some datas concerning the neutron ba-

lance of the thermal burner reactor are given on the Table 5.
23

It is worh mentioning the possibility of a neutron coupled fast/
thermal system which would help to solve the problem should the
thermal burner reactor have too small a k_ due to the large
amounts of neutron absorbing fission products present. In redu-
cing the leakage from the thermal reactor it could be that neu-
tron coupling will improve the entire neutron balance raising the

system breeding ratio over 1.1. (Fig. 8)
24

Fig. 6 Core Volume Composition (VolZ%)

Mo 134

Moderator
BeF2

Thermal "element”
specific power:

o KW
cm” core
~2.67 Sp——
4 em” fuel
Fuel + moderator solu-
¢ tion: ,
/\ B . —_ —
Cp = 1.35 Jg 'K .
= _ C 4 = -5
1V = 2.5 m.s L= 5.57 gem
D - -3
(A o CgOl: 4,55 Jem jh -
o)
> g AT(TC)=950-800=150
<T
A .
9 C%OlAT = 680 Jem 5
O

800°¢ [ime 1ncore,

[l\
T t = 0.255s

I

Fuel velocity
2.5 ms~

fission products

Thermal Reactor Burner
Recycling of

Coolant

Fig. 7

Not to scale

8}
] 1" o
o b »
G =z =l
— e~ Y = =1
T OO O o
Lo 2 OWn QN
O 00 OJN .
STl e) | —
=
o]
el
ad
o
—~ faey
| —_— )
2 W O
[QV] o O
] - [QV]
S = —
O @} O
L . O
ma o &
v — —--—r H
@ ANAN SN N\ @-
=T moy "
- SN SNENTIN N
\ v/.,j - Coopman) — “
D AO.
Q, >
; NENAN AN N i
m \ % ‘/|. /n el ——— ]
[ N | NI N N
-~ il % P (UG ER ER eI NAB M @_ - DR g E
// 4/ 4/ - SN ,,/ ~ NN ~
@ v %q g e — I - .M > e )
P — D o
~ ¢ ) — :
‘/ N\fi P ~aam— ——— “ N N % N N //
VA_ AN t ‘j TS b,/
/\h v N N ™y N N Y
™~ ! ~
A\ X e —— = | o~
1 AN X
— R q \VU ALY AY N )
- A - ]
‘ “ N N //// // // N // //v///
s - AN AN NANANANANANAN AN N NNANENANAYANANANANAN //
N
< — // < T < < < < T < S \ N T ,V/
A‘O\/ RN NG NN N h SN N VA N \
- ——
' q/ N N « N // // N N // N N, ~ N < f/ N // N N N N // // N N // // // I/ // I/ // ~ // // N //

q N N NN NN, N N N N N NN N N RES NN N, N N N N N N N NN NN N S N N N N NN
All N NN N\ N\ N\ AN NN N NN N\ N N N \ NN N AN \ \ N NN\
//r BeTdoY SUTTOL—

26

Table 5

First approximations of the thermal burner reactor neutron balance

Total power: 500 MW
Specific power: MW/1 Volume: 250 litre
Thermal neutron flux: wth = 6 X 1015 n cmmzs_l
2 X 106 p) 3
2 MW/1 gives ~ = 2 x 107 W/cm
1077
Fisslon rate 2 X 103 Xx 3.1 x 1010 fiss/W = 6.2 x lOl3 fiss/cm5
. PU-239 6.2 x 10°° 6.2 x 1000
Plutonium atoms: N 3 —>> T G el
cm (7.2 x 10 )x(6 x 1077) 4.3 x .0
= 1.45 x 1019 Atom Pu/cm3
For 250 litres: NPu—239 (1.45 x 1019)x(2.5 X 105) = 3.6 X 102” =
tot

= 6 Mol Pu-239
Table 5

Amount in Cross Neutron
Volume Fuel steady state Section Balance
(litre) (mol/in core) (octherm,barn) Absorption Production
1 Pu-239/241 6 Mol (80%) Of = 720 4320 36 100
o = 250 1500 | 12.5
Pu-240/242 1.2 Mol (20%) GC = 250 300 2.5
Irradiated Fission products
11 Sr-50 350 Mol 1.20 420 3.5
2.5 Tc-99 57 Mol 22 .0 1260 10.5
7 Cs—-1535 185 Mol 8.7 1600 13.3
1-129 7 Mol A35.0 245 2.0
0.5 :
I-127 14 Mol 6.0 85 0.7
30 Kr-85 68 Mol 1. 120 1.0
Diluent /Moderator
BeF2 5100 Mol 0.01 150 1.25
ZrFu 1100 Mol 0.03 100 0.8
Structure
15 (tubes 7Zr) 1100 Mol 0.3 300 2.5
250 litre Leakage 1600 13.3
Totals 12000] 100% 100

Lc
Fig. 8

coolant

‘ pump

coolant

D

<

s

28

Fast/thermal system

(G

coolant

productecy

G

D

|

Pump fuel

(therm)

pump ‘F&ssion T

fuel 0

B

Rl

e

:1:17 R A
-} ! < e
2 AN \C\<?<>\\\\\\\
5 AR TRNI NN
AN N oo \\
AT AN
7-L_{,;Y- /8/ \\\
)}fiezrlii '/;y/ \\\\\\
1 ‘D’l”,& 11 &) \\
1 AV N
) f p ~§\
‘/[ U £ N
coolant - (e AN N
R
NN N \\\\
Fagt NN
A W Nase NIN NGOV NN )
SN NN | SN NN NN
\\\_ Blanket o
)
©
>~ E‘
——{E; ©
5
[

29

V. The in-core continuous gas purging

V.a)

General remarks

In this reactor an in-core continuous gas purging of the molten

fuel is postulated which significantly improves the safety of this

reactor 1n an 1in-core accident.

A mixture of hydrogen-helium gas is continously bubbled through

the liquid fuel in the core. The mean dwell time of the gas-bubb-

les needs to be controlled and the mean transport time of the

molten components to these bubbles must also be controlled (e.g.

if speed-up 1is desired-intensive mixing, if delay-local addition

of a further gas stream).

The aim of the gas stripping is as follows:

1)

2)

3)

i)

to remove the volatile fission products which in the case
of an accident control the environmental hazard. (I-131,
Xe-133%, Kr-85 and precusors of Cs-137 and at the same
time for the thermal reactor, removal of the I-135, pre-

cursor of Xe-135, improves the neutron balance.

to control the production of delayed neutrons since most
of the precursors and nuclides of this group are very

volatile, e.g.: Br-I-isotopes,

removal of oxygen and sulphur, continously (see Chapter
VIII)

in situ control of corrosion problems on structural ma-

terials (see Chapter VIII).

For the sake of a first approximation a gas flux of 30 cm3 per sec.

(normal state) of H2/He is arbitrarily assumed. At 20 bar pressure
30

and with a dwelling time in core of 20 seconds, the gas bubbles
5

will only occupy a fraction of the core equal to 10 - of its volu-
me and has little influence on the criticality, (but the collap-

sing of bubbles results in a positiv criticality coefficient).

The system proposed for continuous removal of the volatile fission
product from the core itself has a retention time of some hundreds
cf seconds only. Each reprocessing mechanlism which operates out

of core is limited by the amount of molten fuel being pumped

from the core to the reprocessing plant. This amount, due to the
high capital cost of the fuel and high operation costs cannot be
greater than that which gives a fuel in-core dwell time of about
one week. Even with a 1 day dwell time, that is, 1f a after one
day the fuel goes through the reprocessing plant, no acceptable
solution to the I-131 problem is obtained since the activity of
this nuclide 1s only diminished by one order of magnitude

(see Fig. 9).

Only a direct in-core removal gives the dwell time in core as

low as some hindreds of seconds.

V.b) Delayed neutrons emitters

The principal question arise out of the fact that some of the
short lived iodine and bromine (perhaps also arsenic, tellurium)

1sotopes are the precursors of the delayed neutrons.
51

Table 6 Precursors of delayed neutrons for Pu-239 fast fission

Group Half 1life Firaction ! Probable
t 1/2 (seconds) % i Nuclide
1 52.75 3.8 Br-87
2 22.79 28.0 I-137, Br-86
3 5.19 21.6 I1-138, Br-89
I 2.09 32.8 T-139, Br-90
5 0.549 10.3
6 0.216 3.5

As other possible nuclides the following can be considered:

As-85; Kr-92,-93; Rb-92,-93,-94; Sr-97,-98; Te-136,-137; Cs-142,-143.
The removal of these delayed-neutron precursors from the core re-

duces the value of B, which is lower for Pu-239 than U-235.

Thus we have a problem of reaching a compromlse between a removal
as rapid as possible of the hazardous I-131, and as long a dwell
time in the core for the delayed neutron precursors: I-140, I-13%9,

I-1%8, I-137 and the appropriate bromine isotopes.

The compromise is given by a quasi-steady state regime in the core
in which the amounts of isotopes present are given as an example

in Fig. 10.
32

Fig. 9 I-131 in CHLOROPHIL

I-131 Radioactive decay only

10

10

10

Extraction O

Extraction 10-3
) -2
Extraction 10
Farmers safe level of I-131
1 min 10 min 1lh 10h hq hoa l¥ 1Qy
| | | | | ]
| | [ lu T l6 ] |8
10 lO2 lO3 10 lO5 10 107 10

Time, second
33

Fig. 10 Removal of I-131 in CHLOROPHIL I
and the problem of the precursors of the delayed Neutrons

2 GW = 6.2 x 1019 fiss/sec.
I-138 (precusor)
Production Rate
t 1/2 = 5s
A = 0.14 s .
N * 69 Atoms/s
1%o
3 y/
H2 extractio /
p ’
&L
Y _3 a
Ex = 10 x10
= Tx10 At /s
= 0.7% of I-13
Neutron loss = 0.7%

I-131 P = 1xlOl8 Atom/sec.

.
= 101) dis/sec.
a = 30k Curi
~— /3/{> 3 urie

One thousand times smallel
I-131 activity than in
stationary case without
purging

2 1.65 mili Mol

2

Without extraction: after approx 2 months 50 Mega curies I-131

(lxlOl8 dis/sec)
34

In this case the mean dwell time of iodine in the steady state
reactor is about 100 seconds. From the curve in Fig. 9 1t can be
seen that the activity of iodine for a 2.5 GW(t) reactor is of
the order of only 10 k curies (for 100 seconds) instead of

~ 100 M curies in the steady state, a decrease of activity of

5

approx. ].OLl (or 10° for 1000 second extraction rate).

V. ¢) Gas-extraction and other fission products

The gas-extraction also influence the other volatile nuclides.,.
From a very rough estimation for these molten salts (with a
small excess of free hydrogen) the following fission products
and their associated precursors of ilodine and bromine can be

volatile at ~ 1000 °c.

In elementary form: Xe, Kr, Te (2)

In simple volatile hydrides: Bri, IH

In simple volatile chlorides: Sang, SbClB, NbClB, CdClg.
This amount of finally veoclatile components including 1, Br, Xe,
Kr amounts to approximately halfl the total fission products (i.e.
100 micromoles per second). In addition there 1s tne corresponding
amount of tritium (from ternary fission). This amount of all fis-
5/s
or 10 times smaller than the postulated amount of hydrogen f{low
at 30 cm5
197%; Beattie, 1973).

sion products corresponds to a gas volume ratic of about 2 cm

/s . For volatile fission products release: see (I'armer,
55

The extraction removes all short 1lived fission products which are
volatile under these conditions. Thus not only is the removal of
the iodine isotopes and the consequent reduction 1n production

of xenon (e.g. for the atom number: A = 135, 136, 137, 138, 139)
achieved but it slows down the in-core production of Cs-137,
Cs-138, Cs-139 and then also Barium-139 (Fig. 11).

The gas extraction will also cause the vaporization of the higher

components of the liquid fuel: PuCl3 and NaCl.

The fuel consists of:

- 15 mol% PuCl boiling point 2040°K

33
-~ 85 mol% WaCl ; boiling point 1738°k

(see Fig. 12).

One can as a first approximation say that it would have to follo-

wing composition in the wvapour phase: 5 mol% PuCl, - 95 mol% NaCl.

5

The order of magnitude of vapour for pure components at a tempera-

ture of about 1250 K is

NaCl ~5 x 107° bar; PuCl, ~10 ' bar

For the PuClB~WaCl system one assumes here a lowering of the

vapour pressure (thermodynamic activity coefficient approx 0.1).

5

At the postulated volumetric flow rate of 30 cm H2 normal) per

second, the vapourized amount of plutonium is given by:

30 cm3 . 10-4 bar - 10_1 = 10_8 mol Pu/s

5

22000 cm”/mol
A

Mass No.

132

133

134

135 -

136 —

137 -

138 -

139 —

36

FPig. 11 Precursors of delayed neutrons

C) Radioactive Nuclide Volatile elements in this

<> Precursor of delayed neutrons fuel
N\ N
@® stable nuclide - B decay (V,Probabl‘/ volatile
direction T
Sb

131 sn N, L)

| | ]
<10 102 103 10% =105

Half life t 1/2 seconds
Pressure mm Hg

760 —

10

very volatile
S|
’—l-
0Q

noble gases

37

12 Vapour pressure of Fission Products and
Fuel compenents (in pure state)

Fuel Temp in
+ the core

The main fission
products are
circled

The main fuel
components

100

i | ! T | T T T T T
1000 1500 2000

Temperature K
38

This amount of plutonium is of the order of lO—u relative to the
amount of plutonium fissioned in the same time (approx 10-4
mol Pu/s). However, it still has to be recovered which makes

the reprocessing unfortunately more complicated.

Last but not least is the in-core gas extraction of two other

elements

- oxygen in the form of H20 :  oxygen from impurities (i.e.
PuOC1l)

- sulphur in the form of st: sulphur from the nuclear reac-

tion:

35 354

Cl (n,p) (see chapter VII)

VI. The neutron transformation of the fission products

VI. Principle of neutron transformations

Some of long lived radioactive nuclides are shown in Fig. 13 and
14,

The aim of this part of the paper is firstly to discuss the possi-
pbility of the steady state nuclear transformation (steady state
burning) in the thermal neutron region, of the larger part of

the long lived fission products. A first approximation for the
solutions of this problem is given here and a further discussion

initiated.
39

Fig. 13 Long lived radioactive wastes

Mega Curies
for USA

Mega Curies for O Burning in CHLOROPHIL I/EIR
Europe
A ¥ Burning in CHLOROPHIL II/EIR
10 —}03 * Nuclide which cannot be burnt here

s-137

103—\ ¥

10 S Vg Cm-244 \
=)
(2]
3.
o«
¥

Activity - Mega Curies

A\
014 cs-135 %

l | T I
0 100 200 300 400

Time from unloading from Reactor (at the year 2000)
Potential Hazard Index

4o

Fig. 14 Postulated hazard index for long lived fission
products and actinides

10%__ Ingestion

10fl__ Sr-90

I
0 10 100 200

Pu-239
19
107 Am 241, oh2
Inhalation
Pu-288

T I I
0 10 100 200

According to
41

The idea of destroying the beta-active long lived radionuclides

is based on the following:

A g~ AE FP = Fission product,
ZFP —5- (Z+1) (stable) . primary
tl/2 ~10 years FP = Fisslion product after
irradiation
(n,y) neutron irradiation in high A = Atomic mass

flux reactor = Atomlic number

Y E = Stable nuclide

(A+1) . (A+1)E
FP - (stable)
2 tl/2 very short (2+1)

In Fig. 15 are given the nuclide transformations under a high

fflux discussed here.

This problem of nuclear transformation of fission and non-fissio-
nable actinides, being the irradiation products, has been discussed
for many years. The most recent remarks and calculatlions con-
cerning this problem made the following claim: "It 1s not practi-
cal to burn fission product wastes in power reactors because the
reutron fluxes are too low. Developing special burner reactors

Wwith the required (thermal) neutron flux of the order of
I
10

n oecm s or burning in the blankets of thermonuclear reac-
tors 1s beyond the limits of current technology. It seems reaso-
nable to extract tnree actinides (Np, Am, Cm, (M.T.)) for separate
storage or for recycling through the reactors that produce them"

(Claiborne, 1972).
42

Also a controlled thermonuclear reactor CTR was used for calcu-
lation of neutron transformations. In those calculations a neutron
flux of 5 x lOl5n c:m.-gs_l was used with remark that this flux
level 1s somewhat higher than that usually associated with CTR
power plants (Wolkenhauer, 1972, 1973). A three year irradiation
was proposed but it would seem that for the most important fission
products the neutron build-up in relation to the radiocactive

decay plays a relatively small role; e.g. for Strontium-90 after

3 years irradiation all neutron reaction products from (n,y) and
(n, 2n) are lower than 10% of those from the radicactive decay
products Sr-90 and for Cs-137 only (n, 2n) products result in

~ L0% of the radiocactive decay products. But (n, 2n) reactions

need a rather hard neutron spectrum (Gore, Leonard, 1973).

VI. b Cross sections and neutron fluxes for nuclear transformations

In this system the following simple evaluations are made

1) the amount of fission pr-ducts coming from both the fast

breeder (2000 MW) and the thermal burner (500 MW)

2) the cores are purged by means of gas (He + H2 stream) for

ecvacuation of some volatile nuclides (Kr, Xe, Br, 1)

3
%) the fuels and blanket materials are continually reprocessed

) the irradiated fission products are continuously {(or perio-
dically) separated for the sake of eliminating the daughter
stable nuclides (e.g. “r-90 and Zr-91 rfrom the decay and

burning of 3Sr-90)

5) the amounts C in steady state (ss) burning are calculated

by the obvious relationship for the i-th nuclide.
43
Fig. 15 Burning of selected Nuclides

140_ Cs-135/137

139 - 94—
Sr-3g0 2x10 y
_ Zr-93
138 : 93
32m
137 ° 92-
136 91 —
135 90—
La Ce
Atomic Number Z Sr Y Zr Nb Mo 7
| ] i | | | ] | |
55 56 57 58 38 39 40 IA 42
103 -
®Long lived fission product
1024 @ sStavle Huclide
OBeta unstable intermediate nuclide
101 A , . .
@Long lived nuclide
100 -
99 —
T¢ Ry Rh

L

i px(3.1 x 1010) fiss.s_lw_l X ni

C = (mol)
%)

(1

i 2
+ o.9.) x (6 x 10
R 3 J) (

= thermal power (watts)

= yield (relative amount per 100 fissioned Pu-239/241 atoms)
radioactive decay constant (s-l)

= c¢ross section (cm_g)

= mean neutron flux (n cm_zs—l)

o Q > 3w
"

appropriate neutron spectrum (j = fast or thermal)

.
il

The most critical values 0§®3 are given in Fig. 16 for some

selected nuclides for thermal and fast neutron fluxes.

The calculation given above results in the following:

Burning in a thermal high flux power reactor # = 6 x 1015 n cm_gs”l

Cs - 1535

Sr - 90 (FPig. 17)

Kr - 85

Te - 99

I - 129/127 (Fig. 18)

) . _ 16 -2 -1

Burning in fast power breeder reactor: @ = ~ 10" n.cm s

Cs = 137 (Fig. 17/1) (other option: only storage

but of core!)

Zr - 93 (Fig. 18/II1)
- 2
CTIONS, BARN (10 2ucm )

OSH-SECTIHE

CR

NonfIssionable

THERMAL
FLUX

17 2 5 18

1016 10

Neutron Flux (n cm_2s-l)
46

A more complicated problem is the need for the separation of the
two caesium isotopes Cs-135 for thermal burning and Cs-137 for
fast burning or out of core storage. From Fig. 11 it can be seen
that the in-core gas extraction takes out the precursors of
Cs-135, that is the long lived volatile I-135 and Xe-135, but the
short lived precursors of Cs-137 remain in the fuel, so the sepa-

ration of these isotopes 1s possible (see Chapter VII),

Among the most dangerous long lived radioactive wastes are some

of the actinides (see Fig. 13, 14).

The problem of the "destruction" of these nuclides: Np-237,
Pu-238, Am-241, Am-243, Cm=-244, Cm-245, is ‘simple’. In any high
flux fast or thermal reactor because of the relatively large
cross sections these actinides are burned up, the problem is
only limited by the present state of the reprocessing technology
of irradiated fuel. In the most obvious agueous processes e.g.
the aqueous phase - organic phase extraction, these elements:
Np, Am, Cm being hydrolysed, because of a relatively low concen-
tration, and in the hydrolysed form are not extractable by many
organic extractants. Of course a more refined aqueous process
could give a significant improvement in the recovery of these
actinides making it possible to recycle in the fuel and eventual-

ly to burn-up (Fig. 20 and 21).

In the fused salt fast-thermal system proposed here the recovery
of the actinides seems much easier and simpler due to the fact

that

1) the reprocessing occurs not in an aqueous but in a molten
salt medium; the hydrolysis of the low concentratilon

actinides is not possible; the recovery 1s very high.
MOL IN CORE

MOL IN CORE

b7

Fig. 17
_— Only radioactive decay
—_—— Burning in the Reactors
1(Cs 137) 1260 Mol, Amounts in steady-state
Fast Flux V = Volume 1n steady-state,
Thermal flux

Oth = 8.7 barn

! ¥ ] T |

.1 1 10 32 100 1000 1 10 32 100 1000
YEARS YEARS
I1(Sr 90) 102 4 1V (Kr 85)
Thermal flux Thermal flux
L. 1,2 barn " ot - 1,8 barn
10
I T I | 1 T | ] | 1

1 1 10 32 100 1000 1 10 32 100 1000

YEARS YEARS
MOL IN CORE

CORE

MOL IN

10

10

YEARS

Fig. 18
J1(Tc99) 102 1 11(1129,1127)
Thermal Flux Thermal Flux
oth = 22 barn
- 1014 N
23000 Mol 4700 Mol
in 100 years in 100 years
3
3]
- mlO -
o
(@]
E
- =2
010
- 10 4 -
| T ] T | S T 1
0.1 1 10 32 100 1000 1 10 32 100 1000
YEARS YEARS
N
e O BETA UNSTABLE
11(Zr 93) I;i/ . )@ LONG-LIVING
7 (Zr ‘ .
Past Flux E‘] BETA-STABLE
Of - 100 mb SEPARATION POSSIBLE
(2)
T 4200 Mol C) CJ []
—————
A
- 2 - O—-O0—0—0—L{
O
+ | ~o—o0—O
=
_ =
< _.O-—‘Q———D
<
- S—0—O
T T | ! ] ! T T T 1
1 1 10 32 100 1000 Sr Y zr Nb Mo

ATOMIC NUMBER
INCORE

MOL,

10

10

10

10

10

Fig. 19
REACTOR:

49

2,5 GW(t).

ao—

FISSION-PRODUCTS IN STEADY-STATE POWER

Cs 137, fast

Sr 90, thermal

Cs 135, thermal

57 Mol

Te 99, thermal

/ S
o
A// I I 129, thermal
//’;;7 o
~ d A 4 A
// S P T
o wu o) £
PERN Y = .
o A O e
© 3 »
0 v
so) 0
| I | 1
1 1 10 32 o 100 1000

TIME OF IRRADIATION (YEARS)
50

i i i 244Cm
Fig. 20 Burning of the long lived alpha
. emitters dc = 0.4 70 Mol
243 rast .
gt2%Y = 8x10 5ncm
242
241 —

240- @ 800 Mol

5 e 0.33
B 8~
239 @ @ % 6000 Mol
se /T
(n,y) S 0.21
. 238 B
238 U @ @ 100 Mol
(n,pn) 51 2.0
h = o Oj (n, ())
1or1 X

ot (D @

U Np Pu Am

MOL IN CORE

10

10

10

10

10

51

Pig. 21

//
O
o
/o
/ ¥
Y
)
B Y S |
QO
)
b
/// o d
L P ju o]
o wn q [
[ & o
(ORI Q e
[ gV % h
] |

10 32 o4 100

TIME OF IRRADIATION (YEARS)

1000

steady-state core

in
52

2) the continuous remotely operated extraction processes and
preparation of the liquid fuel, and the absence of fuel
elements etc. make this recycling of the actinides (very
hazardous alpha and partly neutron emitters) possible

and desirable.

The high grade recovery of the actinides postulated here gives
the steady state results for the fast reactor shown in Fig. 20
and 21. The reason for burning in the fast and not in the ther-
mal reactor is based on the fact that the relation ég/dzh is

greater in the fast reactor than in the thermal.

It must be remembered that the relatively high neutron flux
results not only in the irradiation of the non desirable radio-
active nuclides but also the stable fission products, not to
mention the stable components of the fuel and the structural
material. A very short review of these partial processes 1is

given here

1) chlorine contains two stable isotopes (underlined) (see

Fig. 22).
- 3
559l (n,Yy) 3501 —E—m-36§£ volatile
17 17 years 108 [
- 38
ey () o AT
12 12

but not only (n,y) reactions are important here. Much more impor-

tant is the following reaction:
559l (n,p) 358 B o BSQl
17 16 8.7 days 17

40 - =
Fig. 22 Chlorine Burn-up
é:éHazardous Nuclide

39 —
table Nuclide
Eventualy enriched

38 — chlorine isotope
and product.

37 1 Cross section for
fission neutrons
millibarns

36 — (n,p) (n,a)

35 Cl1 1.6 )
37 Cl 0.24 ?
35 -
34 — Cl(natural)
Cl(natural) 0.7 mb
33__ A5 C1 0.7 mb
37 C1 0.03% mb
32
31 —
P S Cl Ar K
l | l 1 L
30 | r | | t 7
15 16 17 18 9
54

The presence of sulphur influences the chemistry of the molten
fuel.

2) Sodium has only one stable isotope

Na (n,v) 2Ma Lo 2“@_5_ (n,y)

11 11 12

2 26
51\_4_% (nsY) Mg =

—= Mg (n,y) ‘Mg -£= °Tn

This chaln seems to result in increasing amounts of stable

magnesium and in longer time also stable aluminium.

A short summary of the "history" of the long lived radioactive

nuclides proposed in this paper is given in Table 7.
Table 7T

Nuclide

Sr-90

Zr-93

Tc-99

I-129

Cs—-135

Cs=137

Actinides

55

Problems in fission-products management

Separation

from in core gas purging
diluted with H

from in core gas purging after
decay of other Kr isotopes:

from fuel reprocessing:
separation Sr/Ba needed

from fuel reprocessing: the pre-

liminary separation of Y-GQ1,
precursor of Zr-91 i1s needed

Mixture of Zr-93, 94, g5

from fuel reprocessing

from gas purging, after decay
time of I-131, that 1s
~ 160 days

from gas purging, after the
decay of Xe-13%5, that is after
~ 5 days

from fuel reprocessing: separa-
tion of Cs/Rb 1is needed

from fuel reprocessing together
with plutonium; very high effi-
ciency of separation processes
is needed.

Transformation

not possible,
waste only

burning in thermal
reactor (under pressure
or as fluoride com-
pend )

burning in thermal
reactor

burning in fast reac-
tor; not possible 1n
thermal reactor due
to too large a volume.
The radioactive Zr-93
is diluted by Zr-94,
95

partially used as a
dilutent for the
thermal reactor 7ZrP
or as structural ma-
terial & fission pro-
ducts cladding in
thermal reactor.

burned in thermal
reactor

burning in thermal
reactor

burning in thermal
reactor

quasi-burning (storage)
In fast breeder core-
(In solid form) or

out of core

fissioning 1in fast
reactor

VII. Reprocessing

56

The system proposed here is a rather complicated and sophisti-

cated

'chemical machine'!

and not only a nuclear energy thermal

energy transformer as are other classical reactors.

The entire reprocessing scheme includes:

Table 8

Continuou§ Process

Gas extraction

Fast chloride fuel
reprocessing

Thermal fluoride fuel
reprocessing

Fertile chloride,
material reprocessing

Fission products,
mostly fluoride
reprocessing

Scheme of entire reprocessing system

Short description

Hydrogen-helium
gas stream

Reprocessing of the
2 GW(t) fast breeder

Remark: only plutonium!

No uranium!

Reprogessing of the
0.5 GW(t) thermal
only plu-

tonium! No uranium!

Reprocessing of the fast
breeder blanket material

Mostly uranium, small
amounts of plutonium.

Reprocessing of the-
long lived fission
products,
in the thermal high-
flux.

(Fig. 23)

Principle
action

irradiated in

Secondary

removal of I-131
and Xe

separation of
Cs=-135/Cs-137

removal of stable

nuclides from Zr

57

The effective burning of the most important long lived nuclides
needs a rather complicated reprocessing plant with high tempera-
ture pyrochemical separation methods, both in gaseocus and liquid
salt or/and metallic phases. In some cases also a retention volume

is necessary (for spontaneus decay of short-lived nuclides).

This reprocessing philosophy arises from the following assumptions:
The caesium nuclides are very hazardous but there are at least
two isotopes Cs-135 (t 1/2 = 2 x lO6a) and Cs-137 (t 1/2 2 30 a).

The first one has a reasonable cross section from the point of

view of neutron transformation (Oth = 8 barn), the second a very
small cross section in both fast and thermal reactors of = 100 mb
or less, oth = 110 mb). The separation of two flows e.g. Cs-135

for thermal burning and Cs-137 for quasi burning (rather self-

decaying) could help solve this problem.

In this system the continuous in core gas purgling system is expec-
ted to extract some of the volatile fission products especlally
those with. longer half life period. In the case of caesium the

separation of both isotopes is possible due to the different time

chains
20.8 h 9.1 h 2 X 106 year
I-13%5 - Xe-155 - Cs=-135 -
1137 _ 2S5 xe-137 _ 2% ™ (s5-137 . 30 years

———

The extractable volatile and relatively long lived nuclides 1in
short time gas purging are underlined. The Cs-135 could be

extracted in the form of the precursors. Cs-137 1is retained in
the fuel and must be separated by more conventional (but not so

simple) methods.
58

Another similar problem arises in the case of zirconium isotopes
(see Fig. 18/IV) and iodine (Fig. 18/II). Part of the zirconium
could be used as a diluent of the homogenous fuel for the thermal
core (e.g. 20 mol% ZPFM + 80 mol% FeF2) part as component of the
molten fuel for the fast core (?) and part in metallic form as
structural (but radioactive) material when in the final period of
fast reactor development when the Z2r-93 will be reduced to less

than 1 mass % of the total zirconlium

Some further similar questions will have to be discussed, in some
a separation of isotopes is possible by means of rapid in core
gas extraction or slow out of core (liquid phase) separation

process.

A preliminary draft of the possible reprocessing scheme is given

in Fig. 23.

VIII. Molybdenum and corrosion processes

Tne corroslon processes in this type of reactor are of crucial
importance. In the cores of these reactors are approx 950 kg

molybdenum or approx 10'000 mol.

We postulate here, arbitrarily, a relatively high corrosion rate
equivalent to a full or 100% corrosion of the complete amount

of molybdenum 1in approx 10 years. This means a corrosion rate of:

10'000 mol molybdenum 5

= 3 x 10 ° mol/s = 450 um/s

10 years x (3%.1 x 1O7s/year)
Pu

59

Fig. 23
F,P. = FPFISSION PRODUCTS
REPRO = REPROCESSING
SEPA = SEPARATION
"THERMAL" FUEL
.
~
)
- AN
REP
HERMAL, Tc-99
UE Sr-9p —
AN >
P
2 L N >
\
| THERMAq
| CORE | =
rHERMAL | | >
|[BURNER |
. i
Fiss l
I I \ Prod e
| l \\\:—/
r—flr -
BLANKET l v
\ ) Kr,I,5r,Tc,Cs [ o
\\_,_/
N2 AN ) -
N\ >
N—
- L
LANK - —
MATERT
- >
A >
v
SS‘J.B? s ZP_9
~ N Pu
~ 4 +ACTINIDES
A v
URANIUM PLUTONIUM
(DEPLETED) FOR SALE

TRITIUM

STABLE
NUCLIDES

Br
Xe
Te (?)

Xe-

Xe-130

Zr-91
Ru-100

Rb-86

Ba-13%6

Zr-91

RARE EARTH
NOBLE

METALS
SEMINOBLE
METALS

OTHER
ELEMENTS
ALKALI AND
ALKALI EARTH
METALS
)

-1

of formation (KJ.mol

energy

Free

-50

-100+

-150-

-200+

60

MoC

|
1000

Temperature K
61

Thus 30 micromoles molybdenum per second (or approx 3% mg Mo/s).
In addition to this we have molybdenum as fission product. In
the 2000 MW(t) fast reactor the amount of fissioned plutonium

equals:

2000 MW(t) x 1 day x (1.06 g Pu/MWd(t)) % 2150 g Pu/day

or: 25 mg Pu/s: that is; 100 yu mol Pu/s.

The production side of fission products equals 200 u mol/s.
Molybdenum is present at about 18 mol% of fission products which
means a 36 u mol/s, or almost half the amount of corroding ma-
terial. From both sources, corrosion and fissioning the amount
of molybdenum is about 50 u mol/s. The corrosion processes can

go in three directions

1) Chloride reaction Mo + 2 C1 + Me+ +~ MoCl. + Me
met 2 met

2) Oxygen reaction Momet + 02 > M002

3) Tellurium reaction Momet + Te2 -> MoTe2

These and eventually other molybdenum corrosion processes are
potentially dangerous and should when possible be prevented in
the core itself. This possibility exists due to the reaction

with the gaseous hydrogen (see also Chapter IV).

1) MOCl2 + H2 > Momet + 2HCL
2) M002 + H2 - Momet + 2H20
3) MoTe2 + 2H2 - Momet + 2T€H2 - Te2 + 2H2

(see Fig. 24)
62

Penneability

Solubility and

26

met

gas

wae Wo*ww (pJIBPUBAS) WD
2/T- 3 Hlflml (pJIBPUEBIS) ¢ d

— 7.. Q_..
!
= = o =
l |

Temperature

7___4 J._ /.__._ R_u_

o o o o

- . | -
uge woqly oy /woqBR H A3TTTANTOS

¢/ T~
63

In the extreme case the molar amount of hydrogen is twice as big

as that of molybdenum giving 120 u mol H2/S.

Under normal conditions (1 bar, 0°c) this amount of hydrogen is

acceptable:

~120 umol x (2.2 X ZLOLl cmB/mol) X 10_6 = 3.0 em

For further calculations we arbitrarily assume ten times more,

3

approx 30 cm Hz/s. (see also Chapter V)

Here is a major question concerning the reaction of metallic
molybdenum with hydrogen. From about 100 elements there are only

twelve elements which form no chemical binding with hydrogen.

c’lLl d5 d6 d7 d8
5d - Electrons Mn Fe Co
ia - Electrons Mo Te Ru Rh
5d - Electrons W Re Os Ir Pt

Thus molybdenum fulfills the minimum requirements: 1t has no
reaction with hydrogen. Also manganese-cobalt alloys are suitable

for this reactor (chromium and nickel not!).

In addition the solubility of hydrogen in molybdenum is relatively
small. The dependence of the solubility of H2 in Mo-metal on tem-
perature and pressure is known since molybdenum is a possible
constructional material for containing plasma (hydrogen) in fusion
reactors. Also the diffusion of hydrogen through metallic molyb-

denum 1s relatively very small (see Fig. 26).
{form

oU

Fig. 25 CHLORIDES-OXIDES EQUILIBRIUM
DIAGRAM AT 1000K

Ba
Cs .

- 400 °®
CHLORIDES MUCH MORE
STABLE THAN OXIDES
Ca CHLORIDES IN
FQUILIBRIUM
WITH OXIDES
Na
@
‘300‘~ Pr
()
[ ¥
Zr e
U Pu
-2004 T OXIDES MUCH MORI
@ Al STABLE THAN CHLORIDES
Cr .Sl
' il
\ . il
Feg
. ra
-100 1 ta
¢
®
Mo
T ] T T T I T
-100 -200 - 300
aGEOUOE oY TDES (KJ.1/2 mol 10)

form
65

The diffusion rate of hydrogen from the melted fuel to coolant

and blanket (here also UCl, - NaCl) needs to be discussed.

b

One can assume, however, that also this melt with hydrogen

is saturated so that the porosity of the wall (molybdenum) will
play a minor role. The most important: the variation in mechani-
cal properties of the molybdenum caused by the dissolved hydro-

gen.

The problem of the corrosion of molybdenum in chlorine containing
media is particularly complicated by the numerous moclybdenum

i : M .
chlorides 0012, MoClB, MoClu, M0015
Also the oxXygen chlorine systems for molybdenum and some selec-
ted fuel components including hydrogen are rather complex
(Fig. 24).

In the region of the external heat exchanger the main corrosion
process results from the action of gaseous aluminium chloride

(the secondary working agent)

AlCl3 + % Mo - % MoCl2X + AlCl( )
() () &

This reaction has been discussed in earlier publications (Blander
1957) but unfortunately not all the thermodynamic data 1s known,
The stability of the molybdenum chlorines 1is of course strongly
influenced by the concentration of free chlorine and also by

the temperature,.
66

Fig. 27 Iron and Molybdenum burn-up

(Core materials
60 @ — b0,
oy
B H9 Stable Nuclide
59__ Co
454

(:> Beta-
Unstable Nuclide

58
] Fe
58 Al T&A
.19%
57 ‘
Iron
main
reaction
VA
Molybdenum
A = 10712571
n = fission yield
Z
. — —
952r 9)Nb DS Mg
67

A more detailed calculation of metallic molybdenum corrosion in
the aluminium trichloride is needed. These calculations are

very sensitive to the vapour pressure of chlorides. More detailed
calculations of the corrosion in this system have been given

earlier (Taube, Schumacher 1969).

IX. Molybdenum and iron irradiated in a fast high flux reactor
IX a) Nuclear effects

The high neutron flux irradiation causes physical and chemical

changes in structural materials.

Molybdenum 1s a mixture of stable isotopes (Fig. 27). The most
important by-product of neutron irradiation is the Tc-99 beta-
emitter with t 1/2 2 2.1 x lO5 year and which belongs to the

chain (stable nuclides underlined)

Mo-98 (n,y) Mo-99 B o me-g9 B —
239 66 n 2.1 x 10° a
(n,y)
Te-100 B Ru-100
17 s

for approx 1000 kg. Mo in core or about 10'000 moles the Mo-98

gives 2300 mol. The irradiation rate equals:
y'1e=39 - (2.3 x 10°) x (6 x 10°7y x (10 x 10797y «x 1010 Atom/s
pro sec

= 1.2 X 1017 Atom/s
Total Power Plant thermodynamic efficiency

68

Fig. 28 Approximate thermal efficiency using a
turbine cycle with various working fluids

754
70-
Two circuits with
65+ dissociating fluids in
gas turbine 90 bar
60-
e 7 (N,0,, /8O, )+ (AL,C1 /A1C1 )
50-
454
404
354
(%c)
30 T T T T T
400 500 600 700 800 S00

Temperature
69

After 700 hrs the steady state concentration of Mo-99 (t 1/2 =

66 hrs 2 2.4 x 1053: A= 3 X 10'—5 s-l) equals:
17
Mo-99 _ 1.2 x 10 _ 21
steady 5 = 5 x 10 Atom

The activity of the Te-99 (t 1/2 = 2.1 X 105a = 6.2 x 10125:

A E lO_l3 s-l) after 3% years irradiation of 100 kg of molybdenum

in the fast reactor core:

Activity?g_izar) = 1.2 x 1017 Atom/s x (3 x 3.1 x lO7 s/year)
10717
X T 2 3% Curle/l tonne of Mo
3.7 x 10

IX b) Radiolysis of molten chlorides in very high flux of fast

neutron

The nigh flux of neutrons and especially of the fission fragments
results in a high damage in the f{uel material and cladding. In
the type of reactor under discussicn thils problem seems to be

much easier to solve because

1) the molten chlorides arec fluilds with structure that is
with a very low activation energy for recombination of
the radiolysis product and therefore with a very high

recombination velocity.

2) in core there are no structural materials (tube, clalding

ete, )
70

X. Aluminium trichloride as a secondary coolant and working agen®

In this consideration for the power production a non conventional
gas-turbine cycle is postulated: we have chosen aluminium tri-

chloride, AlClB.

Tt is now clear that this chosen secondary coolant-working fluicd
substance has specific additional properties, which make 1t
particularly intersting in the present application. The reason
is the following: it is well known that real progress in 1im-
provement of the thermal efficiency of turbines is limited by

two parameters:

1) for steam turbines, corrosion effects with non austenic
steel above 580°C on the one hand, and on the other hand
the dramatic increase of costs when an austenic steel is

used.

2) for gas turbines (e.g. helium), the operation of the gas-
compressor consumes more tnan half of the energy produced
by the power turbines. A number of methods are available
for overcoming these limitations, one of these proceeds
as follows: the working agent in the turbine, when the
temperature decreases undergoes a chemical reaction which
decreases the volume. This result is most conveniently
obtained when the working substance polymerises at the
lower temperature. Of course, this polymersiation must
be completely reversible, that is to say that, at the
higher temperature, the depolymerisation process 15
effectively complete. One of the most promising working

agent 1s aluminium trichloride, which, at the lower
71

temperature dimerises by the following mechanism:

(2 A1c1,) B2° o~ (Al.c1,)895 L (A1, C1,)t1a
27767 27767 .
mononer dimer dimer
higher temp. lower temp. lower temp.

When the temperature increases a monomerisation process takes
place. Already in 1959 Blander and co-workers discussed the prob-
lem of AlCl3 as working agent. They wrote: "A number of typical

application (including a gas turbine cycle employing AlCl3 as the
working fluild and a binary vapour cycle employing water vapour
for the lower temperature: M.T.) have been considered but 1in none
of these has the aluminium chloride shown outstanding advantages
over more conventional media. However, it is believed that for
some specilal applications it may well prove to have outstanding
advantages, where the characteristics of the other system com-
ponents are such as to make 1t possible to exploit to the fullest

-
the unique characteristics of aluminium chloride" (Blander, 1959).

Much more optimistic conclusions could be seen in the papers of
Krasin and Nesterenko (1972). These authors maintain that the
overall thermal efficiency of a dual cycle with aluminium chlo-
ride as working fluid in the high temperature cycle and N02 in
the low temperature cycle promises to increase up to 60% for

500 °C and 90 ata (see also: Krasin, 1971).

But with aluminium chloride and with nitrogen dioxide the chemi=-
cal industry has considerable experience, including experience
with corrosion problems, in the high-temperature region. Unfor-
tunately there seems to be a lack of data concerning the erosion
and corrosion behaviour of these substances in turbines and on

their behaviour under neutron irradlation in reactor conditions.
72

Aluminium chloride as a working fluid in a condensation turbine

has the following advantages:

1) the efficiency of the power generation is about 30-50%
higher than for the "classical" working fluids; helium
or steam. The quantity of fuel used 1is correspondingly
smaller. For a country in which the sources of uranium an

and plutonium are limited this advantage may be decisive.

2) the waste heat is smaller an the temperature of the hea-
ting water is lower. In a country in which the only sources
of cooling water are the rivers, such an advantage 1is of

importance.

3) the size of the turbine 1s approximately 5 times smaller
than for a steam turbine of the same power. For a direct
cycle (when the reactor coolant agent is the working fluid
in the turbine) when the circult is radiocactive thewsmall

size of turbine is an important factor,

L) all these properties of an aluminium chloride c¢ircuit have
a significant influence 1n reducing the capital cost of

2 power station.

Let us repeat the advantageous properties: decreased fuel con-
sumption, reduction of waste heat, decrease of turbine size. All
these trends promise to give a significant reduction in power

generating costs.

The crucial problem of possible corrosion processes caused by

AlCl3 may be controlled by the small addition of hydrogen.
Acknowledgements

For discussion parts of this paper the author thanks:

J. Ligou (neutronics calculation), Dr. G. Markoczy (heat trans-
fer), Dr. J. Peter (chemistry problems), G. Ullrich (metallurgy),
Dr. G. Seifritz (neutronics). The author is also indebted for
useful discussions and criticism concerning fast breeder molten
chloride reactor to Dr's, Dawson, Long (A.E.R.E. Harwell) and
Smith (A.E.E. Winfrith).

Also Dr. P. Tempus for his discussion, criticism and support

and Prof. H. Gridnicher for his encouragment and support.

Finally Mr. R.W. Stratton for preparation of the text,
Literature

Beattie J.R., A Possible Standard of Risk for
Bell G.D., Large Accidental Releases.

Blander M., Epel L.G., Aluminium Chloride as a Thermody-
Faas A.P., Neuton R.F., namic Working Fluid and Heat Trans-

fer Medium.
ORNL-2677 (1959)

Development of Adequate Risk Standards

Blomeke J.P., Nichols J.P., Managing Radioactive Wastes.
MeClain W., Physics Today, 8 36 (1973)

Cheverton R.D., H.F.I.R. Core Nuclear Design.
sims T.M., ORNL-4621 (1971)

Claiborne H.C., Neutron Induced Transmutation of
’ High level Radioactive Waste,

ORNL-TM-3964 (1972)

Crouch E.A.C., Fission Products Chain

A.E.R.E.-R-7394 (1973)

Farmer F.R., Proceed. Principles and Standards
of Reactor Safety.

TAEA-SM-169/43 (1973)

Gera F., Considerations in the Long Term
Jacobs D.G., Management of High level Active
Wastes.

ORNL-4762 (1972)

IAEA Proceed. Principles and Standards of
Reactor Safety,

TAEA-SM-169/43 (1973)

OECD Report Disposal of Radioactive Waste,

OECD, Nuclear Energy Agency,
Paris 1972
Rosenthal M.W., The Development Status of the Molten
Haubenreich P.N., Salt Breeder Reactor.

Briggs R.B., ORNL-4812 (1972)

Taube M., A Carbide-fuelled Fast Breeder

Schumacher H., Reactor with Molten Chloride as
Fuel Bonding Material and Chloride
Vapour as Coolant.

EIR-Report Nr. 167, (1969)

Taube M., Molten Plutonium Chlorides Fast
Ligou J., Breeder Reactor Cooled by Molten
Uranium Chlorides.
Ann.Nucl.Sci. Eng. (in press).
Taube M., Steady-state burning of fission

products in fast/thermal molten
salt breeding power reactors.

Ann.Nucl.Sci. Eng. (in press).

Walker W.H., Fisslon Products Data, Part I

ALE.C.L.-3037 (1968)

Wolkenhauer W.C., The Controlled Thermonuclear
Reactor as a Fission Product
Burner.

Am.Nucl.Soc. Meeting,
BNWL-SA-4232 (1972)

See also Am.Nucl.Soc. Winter
Meeting 1973, p. 52