. . . AT HIGH ENERGIES a@ is small, fission cross section low. 
Thus breeding ratio is greater than 1.0 but critical mass is high 
Their Problems and Prospects 
sodium-potassium as the coolant. 
Although enough information exists 
today to indicate that a full-scale 
sodium-cooled reactor could be con- 
structed and made to work, no one has 
yet demonstrated that the operation 
of such a system would be feasible 
in a commercial power plant. Systems 
having primary sodium and secondary 
water cooling, compared with systems 
with only water circuits, require heat 
exchanger and piping welds of much 
greater reliability. Liquid-metal sys- 
tem components, pumps, valves, and 
instruments are beginning to appear 
on the commercial market, but they 
need considerably more engineering 
development before they can be used 
with as much confidence as the more 
familiar components used in water 
systems. 
To prevent plugging and corrosion 
in the coolant channels and piping, 
the amount of sodium oxide present 
in the system must be kept small. 
The control of this contaminant over 
a long period of time may prove to 
be a major problem for large sodium 
systems. 
The scarcity of experience with 
sodium systems is summed up in the 
realization that the EBR-I, a 1-Mw 
experimental system, is the only 
unclassified sodium-cooled reactor that 
has been operated anywhere in the 
world to date. 
However, the reactor programs of 
the United States and Great Britain 
should soon correct this situation. 
The Sodium Reactor Experiment, a 
joint venture by the AEC and North 
American Aviation, will go critical the 
first part of this year. The British 
have scheduled the start-up of the 
world’s first fast power reactor at 
Dounreay, Scotland, in 1958. The 
EBR-II, the American counterpart 
of the Dounreay reactor, is to begin 
operation in 1959, and, finally, the 
EFFBR, the first large-scale fast 
power reactor, is planned for 1960. 
The considerable experience obtained 
in operating these large sodium systems 
should answer many of the questions 
about the feasibility of sodium tech- 
nology in industry. 
Fuel Inventory and Reprocessing 
The fast-power-reactor concept also 
implies high fuel inyentory and extra 
fuel-reprocessing costs. The  criti- 
cal mass of fissionable material of 
a fast reactor is larger than the mass 
of a comparable liquid-cooled thermal 
reactor by a factor of two or so. 
Since a large fuel inventory represents 
a large capital investment the material 
efficiency of a reactor (kw of electrical 
output per kg of fuel) is an important 
factor in determining the economic 
feasibility of a reactor. Although 
fast systems have high thermal efh- 
ciency they by nature have low mate- 
rial efficiency so that the over-all or 
“economic” efficiency of a fast system 
could easily be smaller than that of 
a water-cooled thermal system. 
In addition the concentration of 
fissionable material in the meat of a 
fast-reactor fuel element is roughly 
20% as compared with 1 or 2% for a 
thermal-reactor fuel element. Because 
of radiation damage, both kinds of 
elements would have to be removed 
from the reactor for reprocessing after 
not more than a number of fissionable 
atoms equal to about 2% of all the 
atoms (fissionable and nonfissionable) 
present have been burned up (1). This 
amount of burnup would consume 
almost all of the initial fuel charge in a 
thermal element (assuming enough 
breeding to keep the reactor critical) 
but not more than 10% of the initial 
fuel charge present in a fast-reactor 
element. Thus a fuel atom may have 
to be reprocessed more than 10 times 
before it is finally burned in a fast 
reactor. 
These extra reprocessing operations 
will certainly add to the costs of oper- 
ating a fast reactor directly, through 
the costs of the operations themselves 
and, indirectly, through the larger in- 
ventory requirement and_ fuel-loss 
charges associated with reprocessing. 
On the other hand, unlike the 
thermal reactor the fast reactor is 
insensitive to fission-product poisons. 
Fuel burnup in a fast reactor is there- 
fore limited more by radiation damage 
to the fuel than by available excess 
reactivity. Moreover the reprocessing 
need not remove all of the fission 
products and therefore can be much 
cruder. However, it seems unlikely 
that these advantages will outweigh the 
disadvantage of multiple reprocessing. 
A possible way around this difficulty 
is being explored by the AEC in 
125 
