THE SOVIET JOURNAL OF ATOMIC ENERGY NO. 3

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP10-02196R000100090003-2
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 89
• Document Creation Date: December 23, 2016
• Document Release Date: April 3, 2013
• Sequence Number: 3
• Content Type: REPORT

Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
ATOMIla51
Number 3, 1956
.ATOMIC ENGINEERING
-The Soviet Journal of
ATOMIC
ENERGY
IN ENGLISH TRANSLATION
CONSULTANTS BUREAU, INC.
227 WEST 17TH STREET, NEW YORK 11, N. Y.
Declassified and Approved For Release 2013/04/03 : CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
SOME PROBLEMS OF ATOMIC POWER DEVELOPMENT IN THE USSR'
I.V. Kurchatov
The principal directions taken by the development of atomic power in the USSR are explained and some
problems of the physics of water-moderated atomic reactors are discussed..
The power structute of the Soviet Union is very large. We possess various natural power resources. Siberia
possesses extensive and easily accessible coal deposits and good conditions for the establishment of a network of
very powerful hydroelectric' power stations. The rich water resources there provide cheap water power and the
open coal pits provide cheap-electric and thermal power. In the next 15 to 20y ears .we plan to set up a huge
power network in the Anga0:,-kni1ei basin with a capacity of 250 to 300 billions of kilowatt hours annually.
However, the largey trait of the population and industry of the USSR is at present concentrated on the
plains of the European piiition of the country. The cheap water power resources of this region will soon be ex-
hausted and the mining and transportation of coal to great distances is very expensive. In addition, the rapid
growth of our industry,and agriculture require a greatly expanded production of electric and thermal power._
Our present resources will be sufficient for the next few decades, but in the more distant future atomic
energy may prove to be the practically inexhaustible and relatively cheap source that will insure an abundance
of power in the European portion of the USSR.
We shall have the problem of providing atomic power which, at least under the conditions prevailing in
the European part of the Soviet Union, will be economically more advantageous than coal power. It is clear
that 'only huge atomic electric power plants can provide atomic power economically.
Therefore we plan for the very near future atomic power plants which will each produce about 400 to 600
thousand kilowatts in order to accumulate experience in the construction and operation of such plants as well as
in the large-scale production and processing of fuel elements.
The construction and operation of large plants will also enable us to determine which types of installation
will be least harmful or dangerous to the surrounding population. Such data and the economic factors will de-
termlne the type of electric power plants and the scale of atomic power production during the period from 1960
to 1970.
Between 1955 and 1960 the Soviet Union plans to build five experimental atomic power plants. Construc-
tion of these plants will begin at the end of 1958; some will begin operation in 1959 and the remainder in 1960.
The reactors to be installed in two of the plants will use thermal and epithermal neutrons with a water
moderator and coolant. The electric power obtained from one reactor will be 200,000 kilowatts. In conjunction
with each reactor three turbines rated at 70,000 kilowatts will be operated by saturated steam at about 30 atmo-
spheres pressure.
A second type of plant will be built with reactors similar to the reactor of the first atomic power plant of
the USSR Academy of Sdiences (Professor Blokhintsev reported on this plant at the Geneva Conference). The
thermal neutron reactors will have a graphite moderator; heat will be transfered by water and steam. Steam
at 90 atmospheres and superheated to 480-500? will feed turbines with a combined power of 200,000 kilowatts.
A third type of plant will use a heavy water moderated heterogeneous reactor. Heat will be removed by
a circulating gas. Ai the New York national conference in October 1955 Professor Vladimirsky reported on the
*Lecture delivered April 25, 1956 at the British Atomic Research Center at Harwell.
283
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
basic characteristics of this type of reactor, which will produce steam at about 30 atmospheres pressure and at
a temperature of about 400?C to feed turbines with a total output of 200,000 kilowatts.
In addition to these three types of large atomic power plants we shall construct and put into operation
during 1959 and 1960 a few experimental plants each producing 50 to 70 kilowatts of electric power.
These include;
1) A thermal neutron reactor with ordinary water moderator and a turbine operating with slightly radio-
active steam coming directly from the reactor; "
2) A heavy water homogeneous reactor with breeding of nuclear fuel by the Th232 - U238 cycle.
3) A thermal neutron reactor with graphite moderator and sodium coolant;
4) A fast neutron reactor with sodium coolant and breeding of nuclear fuel by the U238 ? PI.1233 cycle.
Completion of this program of experimental atomic power plants will make it possible to choose the best
types and to act discriminately with regard to many problems of reactor physics which have not yet been solved.
We hope that our achievements will be useful to those nations which because of the state of their natural
resources require the immediate development of atomic power.
We shall now discuss some physical problems associated with reactors in which neutrons are slowed down
by water. In the last few years the institute of which I am director has devoted a great deal of time to these
problems. Water-moderated reactors have a high breeding ratio of atomic fuel as well as simple and compact
construction. In our opinion they are likely to provide a large amount of atomic power in the very near future,
The theory of reactors operating with thermal or fast neutrons, except water-moderated reactors in which
special conditions arise because of the strong influence of epithermal neutrons on physical processes, has already
been comparatively well worked out. In a uranium-water lattice a comparatively large fraction of the neutrons
can be absorbed to produce fission in the range from 0.1 to 3-5 ev, i.e., above the thermal energy range and
below the lowest resonance levels of U238.. This fraction, depending on the lattice parameters and burn-up, can
be as high as 80%.
The simplest theoretical study of a reactor in which the epithermal neutrons strongly affect the process of
multiplication was reported by Professor Feinberg to a session of the Academy of Sciences in 1955. He based his
theory on elementary slowing-down theory, neglecting chemical bonds between the protons of the moderator;
this assumption yielded a qualitative determination of the fundamental properties of the reactor, principally in
connection with the large resonance peak of the Pu239' cross section at 0.3 ev.
So long as there is no plutonium in the reactor core the rated multiplication factor kth is almost independent
of the absorption of epithermal neutrons. However, With large uranium burn-up resulting in the accumulation of
a considerable quantity of plutonium it is important to allow for the epithermal neutrons. Despite the decrease
of n of plutonium in the 0.3 ev resonance, the increase in the proportion of epithermal neutrons results in an
increase of koz,.
Let us consider, for example, two lattices with 50 mm spacing and lumps of enriched uranium and a mix-
ture of U238 and Pu.
TABLE 1
Composition of lump
ko)
0.9% u2"+ 99.1% 0"
0.73% P1.1238 + 99.27% U238
Neglecting
epithermal
neutrons
1.079
1.079
Allowing for
epithermal
neutrons
1.047
1.272
As can be seen from Table 1 the capture
of neutrons in the epithermal region is very im-
portant for the attainment of large bum-up of
uranium in a water-moderated reactor.
For plutonium the value of n in the epi-
thermal region, and especially near its lower
limit, is strongly dependent on the energy; it
is thus important to know the neutron spectrum
and, above all, to determine to what extent the
elementary theory corresponds to reality.
For an exact calculation of the neutron spectrum formation in the region which is of interest to us, it is first
necessary to study the mechanism of neutron collisions with protons which are chemically bound in water mole-
cules. Theoretical work by Drozdov and Gorycnov has established the dependence of the elastic and inelastic
vbeclassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
;
? Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
?4,7
285
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
neutron scattering cross sections for molecules of hydrogen and water on neutron energy below 0.5 ev. It was
assumed that the neutrons are scattered by free molecules of hydrogen and water. The rotational and vibrational
levels of the molecules were taken into consideration. Fig. 1 presents some of these results. The experimental
curves are given for comparison.
04081 2
barnsatot,
11111111111 II 1111111 all 1111
MIN I T?um mum
:g
itin 1 ilk svg1111111111E111111
11111111111 III 11D01111?1111111
11111111i1111 11111111i101111111111111
11111111111i111110111111111111111111
111111111111111111111111M111110
Illopuummonimmaile
innigiiimmiiiiiimmmiii
Nam ifinuimmummumi
UI um inimomum
78/ 2 i 7 at 2 f
7 (,.7 'Emev
Fig. 1. Comparison of experimental and calculated
values of the total neutron scattering cross section
in bound hydrogen. 1) Experimental curve for
molecular hydrogen; 2) experimental curve for
water; 3) calculated values obtained by Goiyunov.
Another method of checking the elementary
theory is the experimental determination of the most
important features of the processes which occur in
the .epithermal region and a comparison with calcul-
ated values.
20
crpuz.r9
--a--
040 45 50
55 68 a, mm
2
6
3
N 6
\\4
Fig. 2. Pu239 and U235 cross sections in a subcritical
uranium-water lattice. 1) Triangular subcritical
uranium-water lattice consisting of natural uranium
lumps of 35 mm diameter with spacing a. The size
of the subcritical lattice is much greater than the
neutron migration length, therefore the neutron
spectrum is determined at points that are distant
from the boundary; 2) nickel disk with film of
Pu239; 3) paper disk for determining background;
4) nickel disk with film of U235; 5) copper disks
for shielding from fission fragments; 6) paper disks
for collection of fission fragments; 7) theoretical
curve; 8) experimental points obtained by Stolyarov,
Nikolsky, Katkov and Antsiferov.
Stollyarov,Nikolsky, Katkov and Antsiferov have irradiated targets consisting of thin films of Pu239 and
U235 deposited on a thin nickel backing. The targets were placed inside slots in the lumps of a subcritical
uranium-water lattice ( Fig. 2). From the activity of fission fragments collected on paper disks the ratio of
the Pu z39 and U235 cross sections was determined as a function of the hardness of the neutron spectrum in the
lattice. In Fig. 2 these experimfntal results are compared with Feinberg's theoretical curve. Of course, these
results cannot be considered to be a proof of the ability of the elementary slowing-down theory to explain the
nature of the phenomenon, but they do show that this scheme can be used to estimate the burn-up of uranium
in a water-moderated reactor.
Measurements performed by Barkov and Mukhin to determine the slowing-down length of neutrons from
the energy of the first indium resonance down to thermal energies show that chemical binding has only a small
effect on the slowing-down of neutrons.
Komissarov, Tarabanko and Katkov from the formation of U239 determined experimentally the breeding
ratio of Pu239 for the initial instant of reactor operation, i.e., when there is still no Pun in the core. The ex?
perimental and calculated values of the breeding ratio as a function of the lattice spacing are shown in Fig. 3.
The agreement can be considered satisfactory in this case also.
On the basis of the foregoing considerations Feinberg, Levin, Osmachkin, Novikov and Saulyev carried out
a series of calculations of uranium burn-up in uranium-water lattices using the electronic calculator of Academ-
ician Lebedev. For these calculations they Used the nuclear constants obtained by Mostov, Pevzner and their
collaborators with a mechanical selector and by Spivak, Erozolimsky, Kutikov and others with a graphite prism.
286
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
It is very important in such calculations to allow correctly for neutron absorption in the accumulated PP); The
results of the calculations depend considerably on the values chosen for the nuclear constants and on the assump-
tions made regarding the spectrum of the absorbed neutrons.
Fig. 3. Breeding ratio (BR) of nuclear fuel in a sub-
critical uranium-water lattice. 1) Paper disk for
collection of fission fragments; 2) uranium disk for
subsequent measurement of -activity of U239 ; 3)
paper disks for shielding from fission fragments; 4)
paper disk for determining background; 5) experi-
mental points obtained by Tarabanko, Komissarov
and Katkov; 6) theoretical curVe.
a fuel element which can operate for long periods while
In connection with the possibilities for U238
burn-up great interest attaches to the recirculation
of nuclear fuel, i.e., to successive runs of burning-
up in the uranium-water lattice. There is ground
for believing that by the use of circulating nuclear
fuel in uranium-water lattices a very high degree
of utilization of U238 can be attained. If we keep
in mind the inevitable considerable losses of nuclear
fuel during the chemical and metallurgical pro-
cessing we must conclude that the possibilities for
a uranium-water lattice are similar to those for
fast-neutron breeding reactors. More accurate values
of the constants, a more detailed study of slowing-
down processes and, above all, further knowledge
concerning the operation of water-moderated re-
actors with large accumulations of Pu 239 will make
it possible to arrive at reliable conclusions regarding
this important problem.
In connection with the possibility of achieving
large uranium burn-up ( even in one run) great prac-
tical importance attaches to the task of producing
being irradiated.
We consider baked uranium dioxide, which is stable under irradiation and does not dissolve in hot water,
to be an excellent material for use in the lumps of a uranium-water lattice. Our prolonged experimentation
with the RFT reactor showed that lumps of uranium oxide even when their cladding is not hermetically sealed
function satisfactorily; there is no contamination of the loop by fission products, and the small activity due to
gaseous fission products disappears soon after stoppage of the reactor.
Utilization of the dioxide results in an appreciable diminution of the breeding ratio in a uranium-water
lattice. Therefore we have not ceased to work towards the production of stable lumps of metallic uranium.
After a number of failures a group of Soviet scientists have succeeded in working out the technology of manu-
facturing good uranium metal lumps. With burn-up of 3 kg per ton of uranium no change of shape is observed
in these lumps; therefore it will be quite possible in the future to plan to use metallic uranium in uranium-
water lattices.
In conclusion I would like to discuss the use of ordinary water in a system employing thorium as fuel.
The problem of breeding U233 from thorium in thermal neutron heavy-water-moderated reactors and in
fast neutron reactors has frequently been discussed by Feinberg, Kunegin and Nemirovsky, who are members of
the Institute; their reseiiches showed that through the use of ordinary water as moderator in a Th232 - U233 sys-
tem it is possible to achieve a breeding ratio close to 1.2 and thus attain complete burn-up of the thorium.
This type of reactor would consist of a core of laminar elements containing U233 and a breeding zone
surrounding the core and containing lumps of Th or Th02. The moderator would be ordinary water at 300?C
and about 100 atmospheres pressure flowing between the laminar fuel elements whose heat charge would be
very high.
An estimate of the possibilities for such a system can be obtained by calculating the breeding ratio for
different ratios of the quantities of water, U233 and construction materials. For the latter aluminum, zirconium,
stainless steel etc. were considered. An important property of a thorium system is the fact that n for U233 is
practically constant at ?2.3 in a:very broad energy range.
The breeding ratio BR of nuclear fuel can be determined from the simple formula
287
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
BR r0 _1_ S
where 6 is the neutron loss in water, construction materials and fission products. in the breeding zone and as a
result of the formation of the isotopes U234, u236 and u235.
6 Depends strongly on the neutron spectrum in the reactor as well as on the burn-up of nuclear fuel in
one run.
TABLE 2
n as a Function of Neutron Energy for U233*
Neutron energy range
Thermal
0.15 - 0.5 ev
0.4 - 3 ev
0.6 - 6 ev
2.5 - 25 ev
8 - 130 ev
30 key
140 key
250 key
900 key
11
Neutron source
2.28 + 0.02
2.28+ 0.09
2.24+ 0.05
2.24? 0.05
2.28+ 0.05
2.28 + 0.05
2.25+ 0.07
2.43 + 0.12
2.45-F 0.12
2.60+ 0.12
Spectral ranges isolated by filters
PhOtOneUtronS
I/
*Data obtained from Spivak, Erozolimsky, Dorofeev, Lavrenchik, Kutikov and Dobrynin.
TABLE, 3
ri ? .as a FtinctiOn of Neutron Energy for U235*
Neutron energy range
Neutron source
Thermal
0.15 - 0.5 ev
0.4 - 3 ev
0.6- 6 ev
2.5 - 25 ev
8 - 130 ev
300 key
140 key
250 key
900 key
2.06 ? 0.02
2.06 ? 0.06
1.60 ? 0.04
1.50 ? 0.04
1.52 ? 0.04
1.48 ? 0.04
1.86 ? 0.04
2.12 ? 0.10
2.21 ? 0.15
2.28 ? 0.08
Spectral ranges isolated by filters
VI
Photoneutrons
*Data obtained from Spivak, Erozolimsky, Dorofeev, Layrenchik, Kutikov and Dobrynin.
II ?
Neutron losses in water and construction materials diminish as the amount of water and construction mat-
erials in the lattice is reduced. When a changes from 20 to 1 the neutron losses become quite small. The spec-
trum of neutron absorption in the core is quite hard even at a=20 (the thermal region amounts to only a few per
cent of all neutrons absorbed in the lattice) and at a = 5 it approximates the spectrum of a fast neutron reactor
(Fig. 4). For such values of a Xe235 pdisoning is considerably smaller than in thermal neutron reactors.
A considerable contribution to the size of 6 can come from the absorption of neutrons with the formation
of u234, U235 and U236. For 30% burn-up the amount of U234 accumulated in one run is about 3%. The amount of
U235 accumulated per run is -0.5%. Since the resonance integral of U234 is probably at least 5 times smaller
than the resonance integral of U233, the contribution to 6 from the latter does not exceed 0.005. After many
sticcessive runs for the purpose of consuming the U233 larger quantities of other uranium isotopes will be accumu-
lated. If we consider a stationary state of the system attained after a considerable period of time and take ?I= 1.5
( Table 3) for U235 in the epicadmium neutron spectrum, 6 does not exceed 0.2. It can be reduced by separating
288
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
7 mey 100/rev iarev key 100ev 10cv ley aloi
Fig. 4. Neutron absorption spectrum in Th232 - U2"
system as a function of relative concentrations of
hydrogen and uranium atoms. a) Relative concen-
tration of hydrogen and U233 atoms.
U"3 from the mixture of uranium isotopes. This
method is technologically and economically ad-
vantageous for the additional reason that the quan-
tity of reprocessable uranium obtained by isotope
separation is comparatively small.
Thus the breeding ratio for the U23- Th232
cycle in a reactor using a considerable fraction of
epithermal neutrons can reach values from 1.10 to
1.20 with 3059 burn-up in one run.
For economically profitable nuclear power
we require in addition to a high breeding ratio a
high specific thermal capacity ( per unit weight of
nuclear fuel). By the use of an ordinary water mod-
erator it is possible to achieve a volumetric heat
release rate in the core of from 1000 to 2000 kw/liter
while the heat release rate of the nuclear fuel is
2000 - 5000 kw/kg.
These heat release rates are also character-
istic of fast neutron reactor cores, but one advantage can be pointed out for the type of reactor which we have
considered. The nuclear fuel in it can be "diluted" with a considerable quantity of construction materials such
as aluminum with practically no reduction of the breeding ratio; this simplifies the task of constructing long-
lasting fuel elements. It must also be mentioned that thorium behaves much better in the reactor than does
uranium. Even when a large amount of U2" was accumulated in thorium we observed no instance in which
thorium lumps went out of order nor any of the changes which are well-known in the case of uranium.
In England you are proceeding very cautiously with respect to water systems; partially for this reason
our research in this area seems to be painted in brighter colors than experimental caution would require. As a
? supplement to this report I have the pleasure of submitting more detailed data* and I await your comments.
*Editor's Note: The author refers to articles in this issue, except for articles by D.I. Blokhintsev and V.V.
Vladimirsky, which appeared in Issue No. 1.
289
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
ON THE NUMBER OF NEUTRONS EMITTED BY Pu.2390N FISSION BY THERMAL
AND SUPERTHERMAL NEUTRONS*
V.I. Kalashnikova, V.I. Lebedev, LA. Mikaelian, and M.I. Pevzner
The average number of neutrons emitted on fission of PU233
by thermal and superthermal neutrons ( Ew from 0.15 to 0.5
ev) is measured.
The fast neutrons that arise on the fission of heavy nuclei are emitted, as is well known, from the fission
fragments, and therefore the average number of neutrons, ii, emitted in one event should be determined both
by the excitation energy of the fragments and their mass distribution. The excitation energy of the fragments
Is to some extent dependent on the excitation energy of the nucleus that undergoes fission, although it is clear
that small variations of the latter ( of the order of the separation of the energy levels) should not lead to vari-
ations in the magnitude of v. Neighboring energy levels of the intermediate nucleus may, however, have dif-
ferent spins, and possible different mass distributions for the decay products. This situation could lead to a new
distribution of excitation energy for the fragments, and thus cause a variation in v [1]. In connection with this,
it is interesting to compare the values of the quantities v for various levels of the splitting nuCleus.
Indirect data as to the behavior of v in the resonance region can be obtained by comparing the results of
experiments on the capture cross section, the cross section for splitting, and the average number of neutrotts
emitted in one capture event, ? teff. Such a comparison, carried out for U233, U25 and Pu239 on the basis of
several works [2-4], does not contradict the usually accepted assumption of the constancy of the quantity v in
the region of low energy for the neutrons causing the fission. In reference [5], however, a discrepancy was found
between the calculated value of veff for Pu 239 and the directly measured value of Leff.
Recently there have appeared several communications in the literature referring to works whose goal was
direct observation of the change in v on fission of U233 [6] and Pu2z4 [6-8] by neutrons with energies from 0.01
to 0.5 ev.
As for U233, in the region in which the experiments were performed, no variation in v was observed. The
measurements made on Pu239 gave contradictory results. In references [6] and [7], no variation in v was ob-
served, but in reference [8], it was found that on going from thermal neutrons to neutrons with energies of 0.3 ev,
V decreases by 12%.
In connection with this, we consider it worthwhile to publish the data that we have obtained on this ques?
? don:
We performed a relative measurement of the value of v for fission of PU233 by thermal neutrons and neu-
trons with energies in the interval from 0.15 to 0.5 ev. There is a strong resonance level in PU239 at 0.3 ev, and
the cross section in the thermal region is to a large extent determined by a resonance level for negative energy
values.
The work was carried out on the neutron beam leaving the reflector of the reactor RFT. The energy in-
tervals were separated with the aid of gadolinium and cadmium filters. The thickness of the filters was chosen
*The results of this work were presented in discussions at the Geneva Conference on the Peaceful Uses of
Atomic Energy in August, 1955.
291
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Vr-71
Figure. ? 1) Chamber; 2) BF3 counters; 3) paraffin;
4) reactor shielding; 5) filter.
so that the lowest effective energy limits of the
neutrons should be, respectively, 0.15 and 0.5 ev.
Thus the resonance level at 0.3 ev was essentially
included between these two limiting absorbing
filters.
A fission chamber with a thin layer of PU239
was inserted into the neutron beam. The neutron
detector was composed of BF3 proportional counters
surrounded by paraffin. A schematic diagram of the
set-up is shown in the figure.
The pulses from the neutron detector were connected in coincidence with the fission chamber pulses. The
ratio of the number ?coincidences of fast neutrons with fission fragments to the number of fission events is
given by the quantity mon , where WI is a coefficient determined by the apparatus ( 17 is the effectiveness of the
neutron detector, and w is the solid angle). A more detailed description of the method and the apparatus used
is presented in reference [9].
The measurement of the quantity man was performed in the free neutron beam and in the beam after it
had passed through the gadolinium or the cadmium filter. It turned out that the quantity man remains constant
to within 1%.
The results of the measurements are presented in the table.
TABLE
Filter
Effective nett-
tron energy'
interval, in ev
Number of
fission frag-
ments
Number of
coinciden-
ces
mai
_ .
Gd (0.08 g /cm2 )
Cd ( 0.86 g /cm2)
0.025
> 0.15
>0.5
331584
335616
80320
46132
46828
11272
0.1391 + 0.0008
0.1395+ 0.0009
_
0.140 *0.002
_
On the basis of these results we may conclude that the quantity v is the same for the two resonance
levels of PU239 ( at 0.3 ev and in the negative energy region). In order to evaluate the accuracy of this asser-
tion, however, we must take account of the fact that in the whole region of energy investigated the fission is
caused by both resonance levels ( we may neglect the effect of the 7.4 ev resonance level). If we assume that
there is no interference between the levels, then the final result may be evaluated as good to within 2%.
LITERATURE CITED
[1] A. Bohr,"Collective motion in atomic nuclei"( Report No. 911 presented by Denmark at the Interna?
tional Conference on the Peaceful Uses of Atomic Energy, 1955).
[2] S.Ya.Nikitin, S.I. Sukhoruchkin, K.G. Ignatyev and N.D. Galanina, Session of the Acad. Sci. USSR
on the Peaceful Uses of Atomic Energy, July 1-5, 1955 (Div. Phys.-Math. Sci.), p. 87.
[3] B.G. Erozolimsky, P.E. Spivak, G.A. Dorofeev, and V.N. Lavrenchik, Session of the Acad. Sci. USSR
on the Peaceful Uses of Atomic Energy, July 1-5, 1955 (Div. Phys.-Math.Sci.), p. 397.
[4] V.F. Gerasimov, V.S. Zenkevich, and V.I. Mostovoi, Session of the Acad. Sci. USSR on the Peaceful
Uses of Atomic Energy, July 1-5, 1955 (data presented by V.S. Fursov).
[5] H. Palevsky et al.,"Measurement of capture to fission ratio of U235, U233 and Pu239 by a new method"
( Report No. 587 presented by the USA at the International Conference on the Peaceful Uses of Atomic Energy,
1955).
[6] B.R. Leonard, Jr., E.J. Seppi and.W.J. Friessen, Bull. of Am. Phys. Soc., 1,1)A2 ( 1956).
28`2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
171 JM. Ataclair, H.II. Landon and M. Jacob, Compt. rend. 241, 1935 ( 1955).
[8] R.L. Zimmerman, H. Palevsky and D.J. Hughes, Bull. Am. Phys. Soc. 1, 1, Al ( 1956).
[9] V.I. Kalashnikova, V.P. Zakharova, A.V. Krasnushkin, V.I. Lebedev, and M.I. Pevzner, Session of the
Acad. Sci. USSR on the Peaceful Uses of Atomic Energy, July 1-5, 1955 (Div. Phys.-Math. Sc.), p. 161,
293
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
DETERMINATION OF THE AVERAGE NUMBER OF NEUTRONS Veff EMITTED
IN A SINGLE CAPTURE EVENT BY THE ISOTOPES U2 3 3, U2" AND Pu239
IN THE SUPERTHERMAL REGION OF NEUTRON ENERGIES
P.E. Spivak, B.G. Erozolimsky, G.A. Dorofeev, V.N. Lavrenchik,
I.E. Kutikov and Yu.P. Dobrynin
The variation of veff for the isotopes U233, U235 and Pu239
is measured in the superthermal region of neutron energies.
For U233 Veff remains constant up to an energy of the order
of 100 ev. For PU232 Veff drops by 12% on going from the
thermal spectrum to the spectrum of energies from 0.15
to 0.5 ev, and then remains constant. For U235 Veff remains
constant on going from the thermal spectrum to the spec-
trum of energies from 0.15 to 0.5 ev, and then drops by
18% on going to the spectrum of energies from 8 to 130 ev.
IN
One of the basic parameters of a nuclear chain reaction which is of use in reactor calculations is the
average number of secondary neutrons that are emitted by the splitting nucleus in one capture event,
of
veff v ? , where V is the average number of secondary neutrons emitted per fission event, and of and cra are
the effective cross sections for fission and absorption,respectively.
The values of veff that have been obtained for the isotopes U233;
U2" and Pu239 in several experiments
[1-3] indicate that a chain reaction is possible in any one of these three isotopes.
For extensive breeding of "nuclear fuel" the value of Veff must be greater than two. In the thermal neu-
tron region this condition is realized for the thorium cycle ( for the isotope U233; Veff ? 2 = 0.28). For the plu-
tonium cycle the difference Veff ? 2 is close to zero.
In order to solve the problem of extensive breeding, it is imperative to know the values of vefffor fission-
able isotopes in other neutron energy regions.
In the present article we give a description of the measurement of the variation of Veff for the isotopes
U233, U235 and PU232 for intermediate neutron energies from the thermal region to 100 ev. The various results
of these investigations were stated briefly by the authors in references [4] and [5].
Here we give a complete account of the data concerning the measurement of veff in the superthermal
region of neutron energies which was performed on the reactor RFT [6] by P.E. Spivak, B.G. Erozolimsky? G.A.
Dorofeev, and V.N. Lavrenchik, and of the additions to this work on the measurement of veff in the neutron
energy region from 0.15 to 0.5 ev which were performed by B.G. Erozolimsky, I.E. Kutikov, and UP. Dobrynin
on the reactor VVR [7] ( water-water reactor).
295
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
The measurement method
A diagram of the measuring apparatus is presented in Fig. 1. The neutron beam from the reactor passes
through a filter of gadolinium, cadmium, or boron, which establishes the neutron spectrum, then through an
inlet collimator, and enters the cavity of a graphite prism. Outside the cavity and evenly spaced throughout
the volume of the prism, are located boron neutron counters, which comprise the indicating system.
In the center of the prism is a sample of the fissionable isotope. In this set-up the number of neutrons
recorded Varies both because of absorption of neutrons by the sample and because ofineutrons resulting from'
fission. This variation ifIthe.indicating system can be expressed in the following war:
AN1 = ? FE avki + FEavk2ueff, ( 1)
where F is the total neutron flux passing through the sample, Eav is the macroscopic cross section for neutron
absorption per square centimeter of the sample, averaged over the whole spectrum of the neutron beam, kl is
the indicator efficiency for the primary spectrum of the neutron beam, and k2 is the indicator efficiency for
the fission neutron spectrum.
Aiik
A0VoYRH20142A944e62454!
AW*,*X*R6keiVotA1t,A0V
2,14, AAWArowwWO,
ZWPAC004:26V6VP4MOVA.
ASN'
..1.1!-0049 A4
WIDNUZ
YVIWVI,
6V159ZO
r
,r.501'ffilWif.M1,11
1007 ,,,VAJWAWZ.
IVA?
g0%.AWYA4Li661.gttordWAY
Fig. 1. Schematic cross section of the measuring
apparatus. 1) Shielding of the pile; 2) first col-
limator; 3) second collimator; 4) the sample
being studied in the center of the cavity; 5) pro-
portional counters (BF3); 6) filter and adapter;
7) monitoring fission chamber with a layer of
U235; 8) shielding ( parafin with boron); 9)
shielding for the front wall of the prism ( para-
fin with boron); 10) insert, retractable for
changing the sample; 11) boron-parafin fil-
ter, inserted for the second measurement.
In order to eliininate the determination of
the absolute values of the neutron flux F and the
total cross section Eav from the measurements, a
second measurement was performed with the same
sample surrounded with a spherical boron-paraffin
filter. The dimensions of the filter are chosen so
that the primary neutron spectrum is completely
absorbed. In this case variation in the counting
rate of the indicator AN2 will be caused only by
the neutrons resulting from fission
AN2 = FEavksveff, ( 2)
where F and Eav have the same meaning as in
Equation (1), and k3 is the indicator efficiency for
the fission neutron spectrum in the presence of the
boron-paraffin filter.
Comparing expressions ( 1) and ( 2), we ob-
tain
_Lk 1
veff
.9 la
1 k2 X
where X is the ratio of 6,/s11 to AN2.
( 3)
As will be shown below, the ratio of the coefficients k1 and k2 is easy to determine with sufficient accuracy
with the aid of simple measurements. It is not difficult to see that the direct determination of the ratio of the
coefficients ki and k2 would necessitate measuring the magnitudes of the neutron flux for the primary spectrum
and the fission spectrum.
We may limit ourselves to the relative variation of the coefficient k1 and thus obtain the relative varia-
tion of Veil'. In order to obtain the absolute value of /Jeff, we can normalize the variation to a known value of
veff measured for thermal neutrons. The ratio of veff for the observed spectrum to veff for thermal neutrons
will be of the form
x
ii _k k, 0
veff lq 1__Lx
k2
( 4)
where quantities with the index 0 are the values obtained from measurements on thermal neutrons.
It should be noted that in the derivation of expression (4) it is assumed that there are no effects from the
296
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
scattering of neutrons by the sample, that the magnitude of the flux F incident on the sample remains the same
in measurements with the boron-paraffin filter LS, N2 and without it ANI, and that the primary neutron spectrum
is entirely absorbed in the measurements with the boron-paraffin filter.
Relation ( 4) will be valid, and the value of veff measured by this method will contain no systematic
errors, only if the above conditions are strictly satisfied. Since the measurements of the effects ANI and AN2
are performed on the same sample, the necessity of knowing the value of Eav is avoided, and it becomes pds-
sible to use samples thick enough for practical measurement, since the effect of self-absorption by the sample
NA
cancels out of the expression for X - AN2 . In this it is assumed that the primary neutron spectrum is not sig-
nificantly altered by blocking of the neutrons at resonance energies.
Thus the problem of measuring veil- reduces simply to the measurement of the relative values AN1 ILL
ki
N2 k2
and
Apparatus and control experiments.
1. The graphite prism was 190 x 190 x 190 cm3, and its cavity was 50 x 50 x 50 cm3. The group of boron
counters that are located at distances from 50 to 60 cm from the boundary of the cavity comprise the indicator
system, whose efficiency depends little on the neutron energy. As will be shown later, this property of the sen-
sitivity of the indicating system makes for the best conditions for obtaining accurate results of measurement
for veff. The large dimensions of the cavity were chosen so as to minimize the effect of interaction of the
sample with the neutrons moving in the opposite direction, which arise as a result of slowing down the neutrons
of the initial spectrum in the graphite.
In order to reduce the effect of these return neutrons, in all the measurements in the superthermal regions
the specimens on which the measurements were being performed were surrounded by a cadmium jacket.
The magnitude of the return effect was determined by a special control experiment. Instead of the sam-
ple, an ionization chamber with a U235 layer covered with cadmium was placed in the cavity. The insert at
the rear wall of the prism was then removed, allowing the beam to pass freely through the prism. It was shown
that the fission fragment count for the hardest spectrum changed within the limits of + 1% when the insert was
removed and replaced.
- The indicating system made up of the counters equally distributed in the graphite prism was insensitive
to the effects of neutron scattering by the sample. The independence of the indicator sensitivity from the di-
rection of motion of the neutrons in the cavity was verified with the aid of a thick graphite sample with a large
scattering cross section ( 1.5 g /cm2 of graphite). It was shown that the effect of neutron scattering by such a
sample is no greater than 0.1%.
The boron-paraffin filter was in the form of a sphere of 26 cm diameter, filled with paraffin and boron,
with an internal cavity diameter of 6 cm for mounting the sample and an opening for admitting the neutron
beam.
As was shown by experiment, the wall thickness of the spherical filter ( 10 cm) was sufficient for practic-
ally complete absorption of the neutrons passing through the opening into the spherical cavity of the filter. In
order to verify this, a thick sample of boron (1.5 g /cm2 of boron), whose absorption coefficient is known to
be several times that of any of the samples of uranium and plutonium used, was placed into the spherical filter,
Careful measurement showed that insertion of this sample into the cavity of the filter in the path of the beam
leads to no changes in the counting rate of the indicator system greater than -i- 0.05%.
2. The dimensions and shape of the neutron beam in the prism cavity were established by the following
collimating procedure: a boron carbide plug 25 cm long with a central opening of diameter 25 mm was placed
in the exit channel of the reactor; further along the axis of the beam in the front wall of the prism was placed
an entry collimator, made of a boron carbide cylinder whose diameter was 10 cm and whose length was 60 cm
and having a central channel of 8 mm diameter.
The dimensions of the beam in the center Of the cavity depend to some extent on the neutron energy,
since the effective length of the entry collimator differs for different neutron energies. For thermal neutrons
297
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
the collimating set-up creates a beam diameter no larger than 15 mm at the center of the cavity. For faster
neutrons the collimation is somewhat poorer. Up to energies of the order of 100 ev, however, the beam dia-
meter is no larger than 18-20 mm.
The small beam cross section and, correspondingly, those of the samples were chosen so as to secure
sufficiently great values for the AN/N effect for a limited amount of matter. In addition, reduction of the area
of the sample and the cross section of the beam leads to decrease both in the return effect, and in the leakage
of neutrons through the opening in the boron-paraffin filter and the collimator. The beam cross section, however,
may not be too small, since then the hard to avoid external background and the background due to neutrons
scattered in the entry channel significantly reduce the relative effect AN/N.
As has already been demonstrated, it is very important that the neutron flux F be the same for the meas-
urements both with and without the boron-paraffin filter. For this reason the diameter of the samples and the
entry channel of the boron-paraffin filter were chosen somewhat larger than the dimensions of the beam ( 22 mm).
That the flux remained the same was verified by two control experiments.
An aluminum sample of 22 mm diameter was placed in the position of the fissionable sample and was ac-
tivated with and without theboron-paraffin filter by thermal neutrons and by neutrons that have passed through
cadmium, in which the neutron fluxes were sufficient for activation.
As a result of these measurements, it was established that the activities were the same in both cases
within the limits of + 1%.
The second experiment was performed with the aid of a group of boron counters surrounded by boron-paraf-
fin shielding. These counters recorded only the fission neutron spectrum, and were not sensitive to the super-
thermal neutrons. The counters were placed in a corner of the cavity. Then the relative effects from a fission-
able sample were measured with and without the ' filter and with various filters forming the neutron
spectrum. In this the ratio of the counts in both measurements remained constant within the limits of + 1%.
3. The isotopes were investigated in the form of powdered oxides which were poured into aluminum boxes
and compressed into small 22mm diameter tablets. A large amount of oxygen in the samples causes the neu-
trons to be slowed down, which can alter the results of measurement of lief due to the capture of slow neutrons
by the cadmium jacket. In order to determine the possible error, the fissionable sample was replaced by a
graphite one, whose thickness was 1.6 g' ./cm2, and measurements were taken with and without the cadmium
jacket. A correction was introduced on the basis of the results of this experiment for the slowing down of the
neutrons by the oxygen, and this correction was no larger than 1% of the total effect.
The choice of thickness for the samples is es-
tablished by the magnitude of the relative effect
AN/N that can be measured with sufficient accuracy.
Even for samples of considerable thickness ( 2 g /cm2),
however, the relative effect AN/N for measurements
in the fast neutron spectrum was merely 5%.
4. In order to measure AN with an error of
+ 2%, it is necessary that the neutron count recorded
by indicating system be taken with an error no
greater than+ 0.1%.
The magnitude of the neutron flux incident
on the sample was 5 x 106 cm-2 sec-2 for the hard-
est spectrum.
For an indicating system efficiency pf about
2% the necessary statistical accuracy of measure-
ment was achieved in a relatively short time as a
result of a few series of measurements.
C
CI
C
Fig. 2. Block diagram of the counting circuit.
C) Boron counters; K) cathode repeaters;
Y) fission chamber with a layer of U235; 1,2,3,
4, and 5) amplification channels; 6,7,8,9, and
10) diode discriminators; 11,12) Schmidt cas-
cade Circuits; 13,14) scalers; 15,16) electro-
mechanical counters (1:16,000).
Stability of the counting circuit was achieved
by the use of negative feedi-back in the amplifiers
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
and by supplying it from stabilized sources. In order to account for the instability of the reactor power, and cor-
respondingly of the neutron flux, an ionization chamber with a layer of U235 placed in the beam was used as a
monitor.
A block diagram of the counting circuit is presented in Fig. 2. Pulses from twelve cathode repeaters con-
nected directly to the boron counters entered four amplification channels. Each of the amplification channels
had an amplification factor of 105 with quadruple negative feed-back, embracing all three cascades. After
cut-off by diode discriminators that follow the amplifiers, the pulses entered a common shaping cascade circuit,
after which they entered a scaling circuit with a resolving time of 0.2 ?sec. The total scaling factor was 16,000.
The total resolving time of the whole counting circuit was 1 ?sec; this made it possible to count up to 104 pul-
ses per second without noticeable omissions.
Fig. 3. Construction of the proportional counter
and its position relative to the neutron beam.
1) Target; 2) aluminum cover with openings
covered with mica (1.2 mg/cm2); 3) casing
of the counter; 4) cathode of the counter; 5)
anode of the counter ( a sphere of diameter 0.16
cm); 6) grid covered by a thin collodion film;
7) copper tube for connection to a gas ballast
volume (90% argon and 10% methylal at a
pressure of 30 mm Hg); 8) insulation:
Measurement of the constants of the
apparatus.
1. In order to determine the ratio of the
sensitivity k1 of the indicating system for neutrons
of a given spectrum to the value of this same con-
stant for thermal neutrons 4 a proportional counter
with a thin mica window was placed into the cavity
of the prism: on this counter was placed a target
covered by a thin layer of lithium fluoride (98% Li6)
in such a way that it was in the same place as is
ordinarily occupied by the fissionable sample. First
the number of L16( Nu) fission events induced by
the neutrons of the given spectrum was measured.
Then a boron or lithium absorber was placed at the
same point, and the decrease in the count AN was
measured. The ratio of the effects was of the form
AN 1 Sabsorber
NIA ? Slayer kl,
where 1(1 is the coefficient that we are looking for.
Since 8 ( the counting efficiency of the pro-
portional counter) does not depend on the neutron
energy, and the absorption cross section for the thin
lithium layer ( E layer) and for the absorber ( E ab-
sorber) depend in the same way on the neutron
energy, we can find the variation of k1 by measuring the ratio AN/NLi in various spectra.
In determining the variation of k1 by the above method, it was necessary, in order to secure measurable
effects, to use thick absorbers. Therefore the effect of self-absorption in the thick absorber, which varies with
the energy, leads to an error in measuring the variation of kl.
In order to determine the correction in the variation of ki due to the effect of self-absorption, it became
necessary to measure the dependence of AN on the target thickness for each spectrum. The weight of the target
was chosen so that the effect of self-absorption should be approximately 15%, and the change in this effect on
going over from one spectrum to another should be no greater than 2-3%.
In order to exclude possible errors due to the presence in the samples of impurities whose cross section
differs from the Uv law (rare earths, cadmium), measurements of AN were performed both on boron and lithium.
The results of these measurements turned out to be the same.
It should be noted that the coefficient Ity, which enters into expression ( 4), characterizes the sensitivity of
the indicating system to the absorption of neutrons by the fissionable samples, for which the dependence of the
absorption cross section does not coincide with the corresponding variation for boron and lithium. It is therefore
very important that the dependence of the indicator sensitivity 1(1 on neutron energy not vary too sharply. In
this case the correction to the value of kJ. measured for boron and lithium is insignificant.
299
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Measurements showed that the value of k1 increased by 1% on going from thermal neutrons to those that
have passed through cadmium, and by 2% for those that passed through a 1 g /cm2 thick boron filter; i.e., in
this neutron energy interval, it is nearly constant.
2. It turned out to be impossible, however, to measure yeti with sufficient accuracy in this prism through-
out the whole neutron energy interval, because on going to harder spectra the background ratio became worse
and the counting rate of the indicating system decreased.
Therefore the measurement of the variation of "elf from the thermal neutron spectrum to the spectrum of
neutrons that have passed through the cadmium was performed with the aid of the prism described above, but
the measurement of the rest of the variation of Veff, into the more energetic spectrum, was performed with the
aid of a smaller prism of dimensions 90 x 90 x 90 cm3 with a cavity 30 x 30 x 30 cm3. The variation of k1
measured with this prism is characterized by a sharp drop ( Table 1) on going from thermal neutrons ( 0.025 ev)
to those that have passed through cadmium ( >0.4 ev), whereas the change in k1 on going from these latter
neutrons to those that have passed through a 4 g ,/cm2 boron filter is only 14%, that is, depends weakly on the
energy.
TABLE 1
The Results of Measurements on the Variation of k1 in the Small Prism
Free beam
1
? 1 g /cm2 Cd filter
(1.2 mm)
0.866 + 0.009
B4C filters
O.SJ,/cm B
0.82+ 0.01
1 ./m2 B
0.79 + 0.01
4 g /cm2 B
0.72+ 0.01
In this case the correction to the coefficient k, that accounts for the difference in the variation of the ab-
sorption cross section of the isotopes being investigated and the variation of the absorption cross section for
boron and lithium is about 2% on going from the spectrum of neutrons passing through the cadmium to the spec-
trum of those passing through the 4 g /cm' boron filter. All the control experiments that were performed with
the large prism were repeated with the small one.
3. The decrease in the sensitivity of the indicating system to fission neutrons when the boron-paraffin
filter is placed in the cavity ( k3/k2) was measured with the aid of a thermal neutron converter ? a fissionable
sample placed in the cavity and enclosed by a cadmium hood. The ratio of counts in the indicating system with
the boron-paraffinfilter to the counts without it, that is, k3/k2, turned out to be 0.31+ 0.01 for the large prism
and 0.40+ 0.01 for the small one.
The results of measurement.
In measuring the effects ANI, AN2, the role of impurities due to other isotopes in the samples being inves-
tigated were accounted for. The PU239 samples contained about 1.5% of the isotope Pu243, which has a strong
absorption resonance at 1.06 ev, and the U235 samples contained about 15% of the isotope U238. In this connection,
when the effects AN1 and 6,N2 were measured for PU239, in addition to the filter that shapes the neutron spectrum,
a filter made of a similar mixture of plutonium isotopes was used to block the resonance neutrons at 1.06 ev.
Similarly, for the U235 samples, an additional filter of U238 was used. The corrections corresponding to this were
from 1 to 4%.
The results of measurement of "elf with all the above corrections taken into account are presented in
Table 2. The first column indicates the material and thickness of the filters, the second gives the effective
neutron energy interval derived by calculation, and the rest give the relative and absolute values of veff for
the isotopes U233, U235 and Pu''. The absolute values of /Jeff are normalized according to the value of veff for
thermal neutrons (1).
From the data presented it follows that ueff for the isotopes U233 and U235 in the energy region of 0.1-0.5 ev
is s the same as Veff for thermal neutrons, and for the isotope Pu235 it is 12% lower than the thermal value. In
the intermediate neutron energy region (behind the cadmium filters) the value of veff for PU238 does not vary.
The decrease of veff for Pu235 in the energy region of 0.1-0.5 ev seems to be connected With the relation be-
tween the radiation and fission widths of the 0.3 ev energy level. For the isotope U235 a sharp drop in Veff is ob-
served for the neutrons that have passed through the cadmium filters, which is probably caused by the properties
300
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
of the set of resonances lying in the region from 1 to 100 ev. For the isotope U233, the value of veff remains
constant throughout the investigated interval.
TABLE 2
Results of Measurement of "elf in the Superthermal Region of Neutron Energies.
Conditions of
measurement
Effective neu-
tron energy
interval in ev
Isotope
Remarks
U233
U235
Pu239
Thermal neutron
beam
0.025
1
2.28+ 0.02
-
1
2.06 + 0.02
-
1
2.03 + 0,02
-
Relative value
Absolute value
Difference of the
effects obtained
with a 1 g /cm2
thick cadmium
filter and a 0.08
g /cm2 thick
' gadolinium filter
0.15-0.5
1.00 + 0.04
2.28 + 0.09
-
1.00+ 0.03
2.06 + 0.06
-
0.88 + 0.03
1.79 + 0.07
-
Relative value
Absolute value
Cadmium filter
1 g /cm2 thick
0.4-3
0.98 + 0.02
2.24-i- 0.05
0.78+ 0.02
-
1.60+0.04
__
0.88 + 0.02
1,79 + 0.04
Relative value
Absolute value
Cadmium and boron
carbide filter 0.5
g /cm2 thick
0.6-6
0.98 + 0.02
-
2.24+ 0.05
-
0.73 + 0.02
-
1.50?. 0 04
_
0.89 + 0,02
-
1.81+ 0.04
Relative value
Absolute value
Cadmium and boron
carbide filter 1
g /cm2 thick
2.5-25
1.00+ 0.025
-
2.28+ 0.05
_
0.70+ 0.02
_
1.52+ 0.04
_
0.88+ 0.02
-
1.79 + 0.04
_
Relative value
Absolute value
Cadmium and boron
carbide filter 2
g /cm2 thick_
8-130
1.00 + 0.025
-
2 28 + 0.05
0.72 + 0.02
-
1.48+ 0.04
-
-
Relative value
Absolute value
The results presented are in good agreement with the work of S.Ya.Nikitin et al. [8], who measured the
variation of veff in this neutron energy region by other methods.
The basic results of the present paper were reported at the Session of the Academy of Sciences USSR on
the Peaceful Uses of Atomic Energy in July of 1955 [51. In August of the same year American and English in-
vestigators published some new data on veff at the Geneva Conference on the Peaceful Uses of Atomic Energy.
Our values for /Jeff for 035 and Pu239 agree with the results of the work of Kanne et al. [9]. In Kanne's
work, the dependence on energy of the ratio of the cross section forracriative capture to'fission cross section;
ar/af is measured. These results make it possible to calculate corresponding values of veff.
1 + a
For the isotope U235 the value of a for a neutron spectrum with an average energy of 100 ev is, according
to Kanne's data, 0.52+ 0.10, which corresponds to veff:= 1.64+ 0.10. According to the direct measurement of
veff in our work, veff = 1.48+ 0,04 for an analogous spectrum. For the isotope Pu239, according to the American
data for a neutron spectrum with a lower bound of 5 ev, veff = 1.82, and according to our results for a similar
spectrum, veff = 1.79+ 0.04.
We should also mention the results of measurements by two groups in the Brookhaven and Hanford labora-
tories [10], in which the variation of Veff is measured on a mechanical selector for the isotopes U2" and Pu239
301
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
in the neutron energy region of 0.1 ev and for the isotope U25 in the region of 0.5 ev. The drop of Veff which
we discovered for Pu233 behind the gadolinium filter, and the constancy of veff for U233 are entirely confirmed
by this work. The average values of Veff for U2" in the energy interval of 0.1-0.5 ev are in satisfactory agree-
ment with our data for that part of the spectrum from 0.15 to 0.5 ev.
Values of veff for the isotopes U233 and U235 have been obtained by a group of English scientists [11],
which also agree well with out data.
Further investigations of the dependence of Leff on neutron energy in the region of several hundred kilo-
electron volts is of some practical interest for construction of fast neutron reactors. The authors have carried
out work on measurements of veff for the isotopes U233, U235 and PU233 in the neutron energy interval from 30
key to 1 mev, the results of which will be presented in the following article.
LITERATURE CITED
[1] P.E. Spivak, B.G. Erozolimsky, "Physical investigations", (Reports of the Soviet Delegation at the
International Conference on the Peaceful Uses of Atomic Energy) Acad. Sci. USSR Press, 1955, p. 184.
[2] A.I. Alikhanov, V.V. Vladimirsky, and S.Ya. Nikitin, loc. cit., p. 199.
[3] D.J. Hughes and J.A. Harvey, Neutron Cross Sections, McGraw-Hill Book Company, New York ( 1955).
[4] P.E. Spivak, B.G. Erozolimsky, G.A. Dorofeev and V.N. Lavrenchik, "Physical investigations", (Re-
ports of the Soviet Delegation at the International Conference on the Peaceful Uses of Atomic Energy) Acad.
Sci. USSR Press, 19.55, p. 213.
[5] B.G. Erozolimsky, Session of the Academy of Sciences USSR on the Peaceful Uses of Atomic Energy,
July 1-5, 1955 (Div. Phys.-Math. Sci.), Acad. Sc!. USSR Press, 1955, p. 369.
[6] "Reactor construction and reactor theory", (Reports of the Soviet Delegation at the International
Conference on the Peaceful Uses-of Atomic Energy) Acad. Sci. USSR Press, 1955, p. 49.
[7] Ibid., p. 91.
[8] S.Ya. Nikitin, S.I. Sukhoruchkin, K.G. Ignatyev, N.D. Galanina, P.A. Krupchitsky and V.F. Belkin,
Session of the Academy of Sciences USSR on the Peaceful Uses of Atomic Energy, July 1-5, 1955 (Div. Phys.-
Math. Sc!.), Acad. Sc!. USSR Press, 1955, p. 87.
[9] W.R. Kanne, H.B. Stewart and F.A. White, "Capture-to-fission ratio of Pu233 and U235 for intermediate
energy neutrons," (Report No. 595 presented by the USA at the International Conference on the Peaceful Uses
of Atomic Energy; 1955).
[10] H. Palevsky et al., "Measurement of capture to fission ratio of U235, U233 and Pu233 by a new method,"
(Report No. 587 presented by the USA at the International Conference on the Peaceful Uses of Atomic Energy,
1955).
[11] P.A. Egelstaff, J.E. Sanders, "Neutron yields from fissile nuclei," (Report No. 425 presented by England
at the International Conference on the Peaceful Uses of Atomic Energy, 1955).
302
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
THE AVERAGE NUMBER OF NEUTRONS ileff EMITTED BY THE ISOTOPES
U233, U235 AND Pu239 ON CAPTURE OF NEUTRONS WITH ENERGIES FROM
30 TO 900 key*
P. E. Spivak, B'. G. Erozolimsky? G. A. Dorofeev,
V. N. Lavrenchik, I. E. Kutikov, Yu. P. Dobrynin
Measurements are taken on veff for the isotopes U233 035 and Pir289 for neutrons
with energies from 30 to 900 key. It is discovered that in this energy region veff
Increases substantially as the neutron energy increases.
Recently several articles [1-10] devoted to the measurement of the number veff of secondary fission
neutrons per absorbed neutron have appeared in print.
In these works, the values of veff are measured for the isotopes U233, U235 and PU239 for slow neutrons in the
energy region from the thermal spectrum to a few key.
In the present article we describe measurements of yen, for the same isotopes for neutrons in the energy
Interval from 30 to 900 key.
The determination of veff is of great interest for calculations of reactors without a moderator.
Measurement of the absolute values of veff for neutron energies of 30, 140, and
250 key
1. Method of measurement
A method for the direct measurement of the quantity veff for fissionable isotopes and thermal and super-
thermal neutrons with the aid of a graphite prism was developed in our laboratory and described previously pi,
[4], [101.
Further extension of this method to two indicating systems, whose efficiencies depend differently on the
neutron energies, made possible its application to the measurements described.below, which were carried out with
the aid of photoneutron sources.
The indicating systems consisted of two groups of BF3-counters located in the graphite prism, with a central
cavity into which could be placed the spherical samples and neutron sources (Fig. 1).
The group of counters that were located in the cavity had a higher sensitivity to neutrons from the source
than to neutrons of the fission spectrum, whereas the group of counters located at distances of about 50 cm from
the cavity boundaries had a higher sensitivity to the fission neutrons. The difference in the sensitivities of the
two groups of counters was made greater by the intro'cluction of cadmium rods into the prism.
" Part of the results of this work have been communicated at discussions at the Geneva Conference on the
Peaceful Uses of Atomic Energy in August, 1955.
303
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
The measuring procedure consisted of introducing a photoneutron source into the cavity and of recording
the number of pulses in both groups of counters simultaneously. The source was then surrounded by a spherical
sample of the isotope being investigated, and the measurement was repeated.
The indications of each counter group (AN' and AN" respectively) which are caused by absorption of
primary neutrons by the sample and the creation of fission neutrons, can be written in the following way;
AAT = ?F/c; Ee Fic; veil.. EAV
ANff Fk; E0 Fk; vqf Loy
(2)
where kJ! and kJ." are the indicator system efficiencies to the primary neutrons, ki and k2" are the same
efficiencies to the fission neutrons, F is the total primary neutron flux passing through the sample, and zav is
the average absorption cross section of the sample for the primary neutrons.
From Expressions (1) and (2) it follows that
yeff
1?aX
(3)
7
Fig. 1. Schematic cross section of the measuring
apparatus.
1) BFrcounters of the "outer" system; 2) cadmium
rods; 3) BF3-counters of the "inner" system; 4) sample;
6) graphite insert; 6) photoneutron source; 7) tube of
the mechanism for inserting the source; 8) pocket for
holding the source.
TABLE 1
Source
Half-life
Neutron
energy
in key
Neutron yield in
number of neutrons
per second
Sb124 + Be
Gan + D20
Na24 + D20
60 Days
14.3 Hours
14.8 Hours
30
140
250
3i0
0.8. 107
2 . 107
.re
304
AN'
Thus, except for the ratio -- = X , only
AN"
the constants for the apparatus a 8, and y, which
characterize the relative efficiencies of the two
indicators to primary and fission neutrons enter
Expression (3).
2. ?Neutron sources
In the measurements, photoneutron sources
composed of Sb 124 + Be, Ga U,+ D20, and Nall + D20
were used; data regarding these is presented in
Table 1.
Gamma emitters were obtained by the
activation of the corresponding isotopes in the
center of the RFT reactor. The high activity of
the sources made it necessary to take serious
precautionary measures in carrying out the
experiments.
3. Measurement of the effects AN'
and AN"
The relative change of the counting rate
AN
?N caused by the hollow spherical sample
weighing about 200 g was only a few percent.
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Therefore in order to obtain the value of X with sufficient accuracy, it was necessary to achieve the possibility
of measuring the indicating system counting rate, accurately to within some hundredths of a percent.
The stability of the counting system was achieved by the use of feedback in the amplifiers and by
stabilized power sources. In order to obtain numerical values for the magnitude of the effects (AN' and AN")
due to the sample being investigated, the indicating system counting rate before the sample was placed in the
cavity was averaged with the counting rate after it had been removed, and then this arithmetic average was
subtracted from the counting rate with the sample in position.
Each of these three successive measurements was taken for 12 minutes, with 3 minute intervals for
inserting or removing the sample. For this method of measurement, the correction for radioactive decay of the
source was no greater than 0.25% of the value of X.
The fact that AN' and AN" were measured simultaneously made it easier to use rapidly decaying sources.
Measurement of the constants of the apparatus
In order to determine the constant of the apparatus a, a photoneution source was placed into the center of
the cavity and the counting rate in both indicating systems was measured. The ratio of these quantities gives
the value of the constant a =
k2"
The constant 8 = was determined in a similar way. For this, the graphite prism was placed close to
the reactor in such away that one of its beams entered through a collimator into the cavity. In the center of
the cavity and in the beam was placed a convertor (sample of U235) used as a fission neutron source. The
convertor was covered with a cadmium sphere with an opening for admitting the thermal neutron beam. The
magnitude of the constant y was determined as the ratio of the counting rates of the inner indicating system
taken first with the corresponding photoneutron source and then with the fission neutron source. The ratio of the
output of these sources was determined in the graphite prism by a method developed earlier in our laboratory [111
TABLE 2
30 Kvv
140 Kev
250 Key
a
0,192+0,0005
0.131+0.00050,134+0,0005
fi
0,254+0.001
0.181+0,001 0,175+0,001
y
1,19 +0.01
1.26 ?0,01 1,22 +0.01
U233
3,22 +0.017
4.42 ?0,04 4.78 ?0.05
X
1J235
2.815+0,015
4.06?0.05 4.55?0.06
Pu239
3,06 +0,014
4.40?0,05 4.85 ?0.06
In Table 2-we present the values of the
constants a, B. y as well as X, for various
neutron energies. On going from one neutron
energy to another, the distribution of the counters
in the measuring apparatus was somewhat changed
In order to secure sufficient difference in the values
of a and B.
5. Control experiments and corrections
a) The scattering of neutrons within the
samples may cause additional effects which were
not accounted for in the derivation of Expression
(3). Errors in measurement related to elastic scattering of neutrons were determined in the following way. A
graphite diffusor (no 2) was placed into.the center of the cavity next to the source. The counting .rates of
the indicating systems did not change by more than 0.5%in.the process. If we take into account the geometry
of the experiment and the thickness of the samples used, it turns out that the experimental error due to elastic
scattering of neutrons is no greater than 0.1-0.20/0.
Inelastic scattering of neutrons in the sample may introduce errors in.the results of the measurement by
changing the neutron spectrum and therefore the efficiency with which the neutrons are recorded by the
indicating system. The influence of inelastic scattering on the accuracy of the measurement was studied with
the aid of a sample of the isotope U238, whose inelastic scattering cross section, as is well known, is much
greater than that for the isotope's tJ, TIN and Pt?". The relative effects -- 'coming' from a spherical
sample of U238, measured with a source made of Na 24 + D20(E0 = 250 key), turned out to be the same in both
the outer and inner indicating systems within the experimental accuracy, which indicates the absence of
effects, from inelastic scattering.
305
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
b) The neutrons from the source that is placed into the cavity, in slowing down in the graphite, create
a field of slowed-down neutrons in the cavity. In order to decrease the effects of interaction of the sample with
these neutrons, all the samples were located inside a cadmium hood, and the dimensions of the cavity were
chosen sufficiently large in comparison with the dimensions of the samples.
In addition, for the measurements "without the sample" the sample being investigated was moved a
distance of about 25 cm away from the source, but remained in the cavity. In this case the direct interaction
of the neutrons from the source with the sample was negligibly small, and the sample interacted only with the
slowed-down neutrons, whose field was practically uniform within the prism cavity. Thus the effect of the
interaction with the slowed-down neutrons,which was no greater. than 4% of the basic effect, was automatically
excluded in the measurement of AN' and AN".
(,e) The fission neutrons created within the sample pass through some distance within the sample itself,
giving rise, in turn, to other fission neutrons. In a thick sample this multiplicative effect leads to a significant
contribution, which makes it necessary to introduce a corresponding correction. The neutron multiplication
factor is determined experimentally. To do this, a spherical source which reproduces the fission neutron
spectrum is placed in the center of the cavity, and the counting rate No in the indicating system is recorded.
Then the source is surrounded by a fissionable sample and the measurement is repeated.
AN
The relative effect ? due to the sample is a measure of the fission neutron multiplication. This
No
effect, measured for the central geometry, was converted to a distribution of primary fission neutrons throughout
the volume of the sample. The problem reduced to finding the ratio ? , where L is the mean free path in the
6
sample of the neutrons created in the sample, and?s the mean free path in the sample of the neutrons
emitted by the source located in the center of the sample. For the samples and sources used, the conversion
factor turned out to be 1.92.
As a result of measurement, the following neutron multiplication factors ? = 1.92 -- + 1 were
No
found for the samples studied;
U233 - 1,10
+ 0.01,
U235-
1,05
+ 0.01,
Pu239 ---
140
? 0.10.
In calculating lieff, the value obtained by Formula (3) must be divided by the corresponding value of gr.
The results of measurements of !Jeff at neutron energies of 30, 140, and 250 key are presented in Table 3.
Measurement of the absolute values of veff for neutron energies ef 250 and 900 key
1. It is impossible, with the above apparatus, to carry out measurements of veff for higher primary
neutron energies, since as the energy increases, the difference in the efficiency of the indicating system to the
primary neutrons and the fission neutrons decreases, causing the accuracy to drop sharply.
In addition, and this is even more important, inelastic scattering of neutrons by the sample begins to have
effect. Therefore, in order to determine veff for energies of 900 key, it necessary to use a different pair of
indicating systems which could avoid the above mentioned complications.
In this variation, a threshold ionization chamber with a layer of U238 placed in the center of the prism
cavity was used (Fig, 2). The other indicator was a group of BF3-counters located at a distance of about 50 cm
from the boundaries of the cavity. The efficiency of this system was therefore practically independent of the
neutron energy; this was verified with the aid of calibrated sources. Thus in this case both indicating .systems
were shown to be insensitive to the effects of inelastic scattering of the primary neutrons.
306
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
TABLE ?3
The dependence of Veff on neutron energy for the isotopes U233, 1.1233 and
Pu239 in the neutron energy interval from thermal energies to 900 key.
Neutron energy
interval
U288
U235
Pti239
Thermal neutron
beam
0,15-0,5 ev
2,28+0,02
2.2810.09
2.06+0.02
2.06+0.06
2.0310, 02 (11
1.79+0,07 1101
0.4-3 ?
2,24+0.05
1.6 10.04
1.70+0,0411.01
0.6-6 ?
2,2410.05
1,50-1-0.04
1.81+0,041101
2.5-25
2,28+0.05.
1.521-0.04
1,79+0,041101
8-130 ?
2.28+0.05
1.4810.04
? 1101
Data from the present article
ayMethOd :of two groups, of counters
30 kev - 2,25+0,07 1,86+0,04 . 2,01?0,05*)
140 ? 2,43+0,12 2,12+0,10 2,35+0,12
250 ? 2,45+0,12 2,21+0,15 2,60+0,18
b) Threshold chamber method
250kev 2,46+0,10 2,00+0,10
900 ? 2,60+0,13 2,28+0,08
2,50-1-0,11
2,57+0,12
? G. N. Pleroy and S. M. Polikanov [12], for Pu239 and 30 key,
obtained veff .2.22 ? 0.16.
Fig. 2. The location of the threshold chamber, the
source, and the sample in the prism cavity.
1) Graphite prism; 2) photoneutron.source;
3) fissionable isotope sample; 4) threshold ionization
chamber; 5) cylindrical electrodes with layers of
u238.
The measurement of yap as in the previous
variation, consisted of determining the change in the
counting rate of the indicators when the source was
surrounded by a sample of fissionable isotope.
It follows from the above mentioned propetties of
the indicating systems, that in this case in Equations (1).
and (2) k2' = kJ'. (the efficiency, of the group of counters
in the prism does not depend on the energy) and ki" = 0
(the ionization chamber is insensitive to neutrons with
energies lower than the fission threshold of U238). Thus
the constant a = 0, the constant y = I, and Expression
(3) for yen- takes on the simpler form
veff 71-13x ?
(4)
AN'
2. Measurement of X = was carried out for
neutron energies of 30, 250, and 900 key (Na'l + Be).
By using the value of veff obtained from the
previous measurements for 30 key neutrons, and the
value of X measured by the threshold chamber method
for the same energy, we may calculate the value of the
207
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
constant of the apparatus $ and thereby obtain the absolute value of ;Jeff for 250 and 900 key neutrons. The
values of veff for 250 key obtained by the threshold chamber method, proved to be in agreement, within
the limits of error, with the results of the previous measurements.
In the measurements, the same control experiments were performed for the evaluation of the corrections
as were performed in the previous variation.' In addition, at neutron energies of 900 key, a correction was
introduced (no larger than 1.5 -2%) related to the fact that a small fraction of the neutrons from the source was
recorded by the threshold chamber.
Results of measurenient
The present work completes a series of studies of veff undertaken by our laboratory in the wide energy
interval from thermal energies to 900 key.
Table 3 presents the data on Veff for the whole neutron energy range investigated.
The results of the last work indicate that the values of 'Jeff for the isotopes U233, U235 and Pu2" increase
materially in the fast neutron energy region. This is in agreement with the qualitative results of statistical
theory, according to which the ratio of the radiative capture cross section to the fission cross section decreases
as the energy increases in this energy region.
If we know the variation of the fission cross section [3, 13, 14] in this neutron energy region, we can
make some conclusions as to the variation of the radiative capture cross section. It turns out that the radiative
capture cross section for U2", U235 and Pu239 follows the 1/V law in the neutron energy region from 30 to 900
key. The substantial increase of veff of the fissionable isotopes establishes conditions favorable for the opera-
tion of fast neutron reactors with extensive breeding of nuclear fuel.
The authors take this opportunity to express their gratitude to Academician I. V. Ktuchatov, under whose
initiative the experiments devoted to the study of the dependence of veff on neutron energy were conducted.
LITERATURE CITED
[1] P. E. Spivak, B. G. Erozolimsky, Physical Investigations (Reports of the Soviet Delegation to the
International Conference on the Peaceful Uses of Atomic Energy), Acad. Sci. USSR Press, 1955, p. 184.
[2] A. I. Alikhanov, V. V. Vladimirsky, S Ya. Nikitin, Physical Investigations (Reports of the Soviet
Delegation to the International Conference on the Peaceful Uses of Atomic Energy), Acad. Sci. USSR Press,
1955, p. 199.
[3] D. Hughes, and A. Harvey, Neutron Cross Section, U. S. Atomic Energy Commission, July, 1955.
[4] P. E. Spivak, B. G. Erozolimsky, G. A Dorofeev, V. N. Lavrenchik, Physical Investigations (Reports
of he Soviet Delegation to the International Conference on the Peaceful Uses of Atomic Energy,) Acad. Sci. USSR
Press, 1955, p. 213.
[5] B. G. Erozolimsky, Session of the Academy of Sciences USSR on theltaceful Uses of Atomic Energy,
July 1-5, 1955 (Div. Phys.-Math. Sci.), Acad. Sci. USSR Press, p. 369.
[61 S. Ya. Nikitin, S. I. Sukhoruchkin, K. G. Ignatyey, N. D. Galanina, P. A. Krupchitsky, B. F. Bellcin,
Session of the Academy of Sciences USSR on the Peaceful Uses of Atomic Energy, July 1-5, 1955 (Div. Phys.-
Math. Sci.), Acad. Sci. USSR Press, p. 87.
[7] W. R. Kanne, H. B. Stewart, F. A. White, Capture-to-Fission Ratios of Pu239 and U235 for Intermediate
Energy Neutrons (Report No. 595 Presented by the USA at the International Conference on the Peaceful Uses of
Atomic Energy, 1955).
'Ass
[8] H. Palevsky, Measurement of Capture-to-Fission Ratio of u and Pu239 by a New Method. (Report No.
587 Presented by the USA at the International Conference on the Peaceful Uses of Atomic Energy, 1955).
308
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
[9] P. A Egelstaff, J. E. Sanders, Neutron Yields from Fissile Nuclei (Report No.42 5 Presented by England
at the International Conference on the Peaceful Uses cif Atomic Energy, 1955).
[10] P. E. Spivak, B. G Erozolimsky, G. A. Dorefeev, V. N. Lavrenchik, I. E. Kutikov, Yu. P. Dobrynin,
Atomic Energy 1956, No. 3, 13 (T. p.295)..
[111 P E Spivak, B. G. Erozolimsky (to be published).
[121 G. N. Flerov and C. M. Polikanov, Report Acad. Sci. USSR, 1954.
[13] G. A. Dorofeev and Yu. P. Dobrynin (to be published).
[14] V. I. Shigin, Dissertation, Inst. Phys. Prot). Acad. Sci. USSR, 1954.
* T. p. = Consultants Bureau Translation pagination.
309
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
MEASUREMENT OF THE AVERAGE NUMBER Veff OF NEUTRONS EMITTED
PER SINGLE CAPTURE EVENT FOR SAMPLES OF Pu239 WITH IMPURITIES
OF THE ISOTOPE Pu246 AND THE EFFECTIVE RESONANCE CAPTURE
INTEGRAL OF Pu2"
B. G. Erozolimsky, I. E. ?Kutikov, Yu. P. Dobrynin, M. I. Pevzner
L. S. Danelian, S. S. Moskalev
In this article we present the results of measurement of veff in Fermi spectra
with lower bounds of 0.15 ev (gadolinium filter) and 0.4 ev (cadmium filter) for
samples of PU239 with various concentrations of impurities of the isotope PIP? (0%,
1.6%, 2.5%, 6.5%, 16%) and the effective resonance absorption integral of Pull?.
Introduction
In using an atomic reactor as an energy source, it is economically profitable to make use of long lasting
fuel cycles, that is to operate at conditions of thorough "burning" of the fissionable isotopes. For these
conditions, the neutron balance in a plutonium reactor will depend not only on the properties of Pu239, but
also on the properties of the isotope Pull? that accumulates in the reactor.
A very convenient gauge of the influence of the isotope Pu249 on the neutron balance in a reactor.is
the quantity veff for samples with impurities of one or another concentration of Pu249, which characterizes the
duration of operation of Pu.
When this data is obtained, it makes possible the determination of the effective resonance capture
integral of Pu249.
In the present paper we present the results of measurement of veff in the Fermi spectra with lower
bounds 0.15 ev (gadolinium filter) and 0.4 ev (cadmium filter) for two sets of samples with impurity concen-
trations of the isotope Pu240 of 0%; 1.5%, 2.5%, 6.5%; 16%; also the effective resonance capture integrals are
evaluated.
Analysis of the Pll239 samples for impurity concentrations of the isotope p u2 40
In the operation of a uranium reactor, as a result of successive neutron capture, several plutonium isotopes
are created (Ru239, Pu249, Pu, PU242, etc.).
The half-life for a -decay of Pu249 (6500 years) is less than the half-life for a -decay of Pu299 (24,000
years). Therefore the amount of Pu240 in the sample can be found by determining the specific a -activity of
the sample. But this method gives sufficient accuracy only for the case when thea -activity of Pu243 is a
noticeable part of the total a -activity, and it is therefore not applicable to samples with small concentrations
of Pu249.
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R0001000900031
-12
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
An analysis by amplitude of the a -particle spectrum of plutonium samples is made extremely
difficult by the fact that the basic a -particle energies of PU239 and Pull? differ only by 12 key.
The spontaneous fission half-lives of PU239 and Pu249 differ in a more substantial way, being equal to 1.25?
? 1011 years for Pu2" and 5. 1015 years for Pu239. Clearly the number of spontaneous fission events in a plutonium
sample is proportional to the concentration of Pu249, the part due to Pu239 being negligibly small. Only Pu242
can give a noticeable additional contribution to the number of spontaneous fission events (To = 7.25. 10"
years). The amount of PU242 in our samples is, however, extremely negligible.
The analysis of the isotope composition of plutonium samples by measuring the number of spontaneous
fission events in an ionization chamber is promising, but involves much work. Therefore,this method was used
to find the percent concentration of Pull? only in one sample. Determination of POW concentrations in the other
samples was carried out relative to this standard sample. The presence of a strong isolated resonance in Pu249
at 1.06 ev energy, and the absence of resonances at this energy in the other plutonium isotopes, makes it
possible to determine the relative concentration of Pu249 in samples from the areas of the dips in the trans-
mission curves obtained on a mechanical neutron neutron chopper.
Determination of the Pt.?'" concentration in the standard sample. The fragments from spontaneous
fission of PJ49 were recorded in a multi-layer ionization chamber. The substance being studied was deposited
on a thin nylon film, which made it possible to record the pulses from both fission fragments simultaneously
and therefore to work at a discrimination level that excluded the pulses contributed by a -particles. The
construction of the ionization chamber and the method of preparing the layers on the nylon films is described
in reference [1].
The amount of substance deposited onto the film was determined by comparing the total a -activity of the
layer on the film with the a -activity of a layer of the same substance whose weight was known, which was
electrolytically deposited on a platinum target." The concentration of PIP in the sample which was chosen
as the standard was 6.4 0.25%.
TABLE 1
Concentration of Pu0 in the samples
Sample
COncenqation of Pu249 'in %
No. 2
1.56 ? 0.12
No. 3
2.5 ? 0.2
No. 4
6.4 ? 0.25
No. 5
15.8 I 1.3
? A control experiment on the determination of the absolute number of a -particles emitted by the layer
on the platinum target gave a value of 6.7 f 0.4%, which is in good agreement with the previous result.
- Determination of the concentration of Pill" in the samples by the transmission curves. As was mentioned,
the presence of a strong neutron resonance in the isotope Pu249 at 1.06 ev energy makes it possible to determine
the concentration of PU249 in the rest of the samples relative to the standard one by the area of the dips in the
transmission curves. For thin samples, the area of the dip is proportional to the number of Pu249 atoms in the
sample. In order that the transmission curves for samples with various percent concentrations of Pu249 not be
measurably different in the resonance region, samples with about the same number of Pull? atoms per cm2
were prepared. In this, previous data from isotope analysis with a mass spectrographwere used. The area of the
dip was determined by the usual method, with corrections for the area under the wings of the transmission
curve [2].
Since, in actuality, the samples studied did not satisfy the conditions for a thin sample (n 10 ev. All this
should lead in too low a value for the experimentally determined value of T, since it corresponds to the slowing
of neutrons to an energy E > 1.46 ev.
The correction to r exp was calculated with the aid of the relation
(b) thermal neutrons, and (c) the function pt(r).
39 ev N9.1 ev
.
T 71.46 ev ?7(3.9-1.46 ev) N r(9.1-1.46 ev) N
,30 key,
c dE
I, r(E-1.46
10 ev
Here N is the total number of neutrons absorbed by the inditim foil, Ni is the contribution .of the particular resonance
XiXs E ev E ev
to the foil's activity,
T(E =1.46 ev) 3 In 1.46 ev ? 0.42 In 1 c is a constant determined by the
. 46 ?
activation cross section in the energy region E >, 10 ev, where we took the cross section to have the dependence
0
2
a ? E 1024 cm2, The magnitude of the correction proved to be 0.2 ? 0.1 cm2.
323
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
The large value for the uncertainty ? 0.1 cm2 is related to the possibly inaccurate determination of
T(E ?1.46 ev)by age theory and to the uncertainty in the variation of the activation cross section in the energy
region 5> 10 ev. Thus the following values are obtained for the square of the slowing-down length to an energy
of 1.46 ev:
71. ev = 586 0.15 cm2 for the large source, and
46 , I
T1. ev = 5.48 + 0.15 cm2 for the small source. 46
In principle it is possible for the indium foil to be activated by the y -rays of the photosource according to the reac-
tionlnis (y , y ') Inns". In order to verify the absence of an error due to this effect, P + Cd(r) was measured
at distances greater than 18 cm. The absence of activation as a result of the y -rays at a distance of 20 cm proves
that it is also absent at small distances, since water absorbs y -rays weakly, whereas the neutron density decreases
rapidly with distance.
In order to determine the thermal neutron density p t(r), a measurement of the dependence of the indium
foil activation (in the steel adapters) on distance from the source was performed. In this case, as opposed to the
measurements of p in cd (r) in the cadmium adapters, the indium foil is activated not only by resonance neutrons,
but also by thermal neutrons and neutrons of "intermediate" energies from 0.025 to 0.5 ev, which are blocked
out by the cadmium in the measurements of p + cd(*. Therefore in order to get the correct value of M2, the
activity caused by the resonance and intermediate neutrons must be excluded from the total activity of the indium.
The activation by the intermediate neutrons can be evaluated by calculation. In fact, the neutron flux activating
the foil is proportional to 1/E. It follows, then, that we may compare the activity of the foil caused by the
intermediate neutrons with the activity caused by the resonance neutrons. The latter is known from the measure-
ments of P in + Cd (r) . The results of the calculation must be normalized to P in + Cd (r).
Such a calculation, however, cannot give an accurate result, because of the fact that in the energy region
E < 1 ev scattering by water molecules begins simultaneously with scattering by free protons. It is possible to
give only two evaluations, above and below the magnitude of the contribution to the activity caused by the
intermediate neutrons. In getting the low value, it was assumed that from E = 0.5 ev up to thermal energies, the
scattering is by free protons only ( = 1, a, =20' 1044 cm2). In getting the high value, it was assumed that in
the energy interval from 0.5 to 0.1 ev the scattering is by free protons, but that from the energy E = 0.1 ev on
2 -24
down, it is by water molecules (g A + 2r ? 0.107, as = 80- 10 cm2). The results are, respectively
(1) -
Pinter - 0'77 Pin + Cd (r),
PC2)
inter= 1.43 P In + cd (r).
Thus M2 should be calculated from the curve
in one case, and from the curve
) p 1.(r) ? 1,77 Pin+cd (1.)
t(25 (r) = p (r) 2,43pini-cd (1)
in the other. The corresponding values for M2 are M24) = 13.92 ? 0.05 cm2 and M2(21 = 14.35 ? 0.05 cm2.
m2 +M2
(1) (2)
2 with an
Since the correct value of M2 is contained between M2(1) and M2 it was taken as M2 -
(2)
m2 m2
uncertainty equal to half the difference of the values,
2 = 0.22 cm2,
= 14.13 ? 0.25 cm2.
29.4
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Additional measurements made with a "point source indicate a small change in the curve for pin(r) at
small distances (dotted line on curve b in Fig. 2). The value of M2 determined from the curve for the
"small source" is M2 = 13.64 0.25 cm2.
Together with T1.46 ev and M2, the diffusion length L for thermal neutrons in water was also determined in
this experiment (t = 19?C). The necessity for measuring L arises from the fact that the experimental values for
the diffusion length that appear in the literature do not agree with each other (see the Table).
For an Sb + Be source the distribution of resonance electrons in water falls so rapidly that from a distance
r 15 cm on, only thermal diffused neutrons exist. Thus for r> 15 cm we may write a diffusion equation without
a source
? '
' V2P.?
1
where p is the neutron flux density.
In our concrete case of a spherically symmetric distribution in a system of coordinates centered at the source,
the solution is of the form
e?r/L
Then L can be found from the slope of the straight line pr e'hi drawn on semi-log paper.
The function pt(i) is drawn on Fig. 2 (curve c).
From the slope of the line it follows that L= 2.68
Diffusion Length of Neutrons in Water 0.02 cm. Additional measurements made in dis-
tilled water gave L = 2.69 0.02 cm. The relatively
high accuracy of the result is due to the large interval
for the measured curve (15-35 cm).
TABLE
L, (cni)
Method
Reference
Comparing the experimental data obtained for
3?0.3
Direct measurement
151
the large source,
2.77+0 .04
*
161
2.76+0.03
171
3.25+0.13
2.67+'0.02
2.72+0.04
Pulse method
[81
191
[10]
i .46 ev= 5.86 ? 0.15 cm2,
14.13 ?0,25 C.442,
2.69+0.08
[11]
2.85+0.05
[12]
L2=7.18? 0,11 c.442
gives, for the quantity r = M2? r1.46 ev ?L2.
the value
AT(1,46-0,025 ev) -=2* 1.1 ? 0,5 cs2.
Comparing the experimentsl results obtained for the small source,
gives, for Ls, r, the value
Ti.46,,= 5.48 ?0,15 c.4c2,
M2=13.64 ? 0.25 c4t2
6'C(1,46-0,025, elf) = 1.0 ? 0,5 c.42.
325
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
It should be noted that a large part of the uncertainty in Ar comes from the large total correction in working
over the results of the measurements with the large and small sources.
The near agreement of the values of Arcbtained from the large and small source is to be expected. Indeed,
in the slowing down of neutrons from an energy of 1.46 ev to 0.025 ev, the resonance neutrons serve as a distri-
buted neutron source; they rapidly slow down to 0.025 ev energy, and then diffuse. In so doing, the diffused
neutrons seldom come to the point where the source is located, and therefore its influence is rather weak.
? In slowing down from 30 key to 1.46 ev, the place from which the neutrons come is the source itself, and
therefore its dimensions influence the value of 7 quite strongly. Comparing the values of 7 obtained with the
large source (.r = 5.86 f 0.15 cm2) and the small one (7. = 5.48 ? 0.15 cm2) , we see that the difference between
them is as large as 0.4 cm2 and that the value of 7 corresponding to a point source must therefore be less than
5.48 cm2. Calculation of 7 by Marshak's formula [13] for neutrons slowed down from 30 key to 1.46 ev gives the
value Ttheor = 4.9 cm2, which is 0.6 cm2 lower than the experimental value for the small source. It is possible
that the fact that the source used is not a point source is one of the reasons for the difference between rexp and
rtheor. A second reason for this difference may be the presence of neutrons of about 300 key in the Sb + Be
source.
An additional verification of the reliability of the measured density of resonance and thermal neutrons can
be obtained from their mutual agreement.. If the curve of the spatial distribution of resonance neutrons is con-
sidered as a distributed neutron source, then we may find the distribution of thermal neutrons by calculation.
With this purpose, the spatial distribution curve of the resonance neutronsPIn + Cd(r) was approximated by the sum
?T 2/4 Bi:
.of three Gaussian curves PIn + Cd (r) =ai e
1
The curve for neutrons slowed down to E = 0.025,ev was found by the formulas
f'?r I
e
PIn+Cd lr'/ 1 r' ? r I
,
Pb1.025 9a, (r) (413)8/ 2 ePi/L,2
where
ai
?r/L' 119 (xi)
Ir'?r12 ?r 12 cc 0L4A, ra
8/2
co.)025 De +
PInCd (r') a AT. dr
4AT ) e
-1212 dt; r)I VA; X21 = r)/1/A ;
' cm; &t=1,2 em2 *).
eriv tJ
al,
* Ar = 1.2 cm2, not 1 cm2 , because the experimental curve of spatial distribution of resonance neutrons is being
used, and the contribution of 0.2 cm2, which is related to the activation of the indium foil by neutrons of energy
E> Er, should not be excluded. ?
326
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
In the first case it is assumed that the spatial distribution of neutrons slowed down from the resonance
energy to the thermal energy is determined by diffusion theory; in the second case, by age theory. Since Pr
and L'2 are small, the difference between the curves of
PO ,1025) ev and po(2.02) 5 ev is also small, and either of the
curves of pr0.025 ev may be used to obtain the thermal
neutron distribution due to diffusion in water:
? */1 t (r
106
10
to?
10'
>a?o
t5 15 30 r CM
11"?r1
e L
Pt(r)= P0.025 (r') , dr'
? 1There L = 2.68 cm is the diffusion length of thermal
neutrons in water. As can be seen from' Fig. 3, the calcu-
i, :,?.c.1 curve agrees well with the experimentally meas-
thcri one. In obtaining M2 , it was assumed that the
neutrons are not thermal until their energy becomes
0.025 ev, and in order to obtain the distribution of purely
thermal neutrons from pin(r), together with the activation
from resonance neutrons, we subtracted out the activation
caused by neutrons with energies E0.025 ev. < E < Er.
Since the number of neutrons with energy E > 0.025 ev
in the Maxwell distribution is large, the question as to the
meaningfulness of this procedure arises. ,It was possible to
pick a nominal limiting energy below which the neutrons'
. would be considered thermal. For instance, we could
have taken 4 x' 0.025ev =0.1ev for this energy. In this
case we would have to subtract only the resonance
neutrons from p In (r) and the quantity Li 1(146 ev ?o.rev)
would be 0.6 ? 0.3 cm2 instead of 1.0 f 0.5 cm2 for
? 410.46 ev ?0.025 ev) Both of the experimental values
can be compared with the calculation according to
M?arshak's formula (or according to age theory), made
under the assumption that the slowing down of neutrons
is achieved by the free hydrogen and oxygen nuclei
and
Fig. 3. Comparison of the calculated curve for p r(r)
(solid line) with the experimental data (points on the
curve).
=
rtheor (1.46 ev-0.025 ev) L 7 cm2.
2
Artheor (1.46 ev-0.1 ev) = 1.1 cm ?
In both cases the experimental values of Pr are somewhat smaller than the calculated ones. The interpre-
tation of the results obtained is complex, since in slowing down neutrons from 1.46 ev to 0.025 ev the effect of
the chemical bond is felt in several factors acting in opposite ways: 1) the increased scattering cross section due
to the fact that the proton is bound to the water molecule; 2) the increase in the symmetry of the scattering, as
a result of which Xr decreases significantly; 3) decreased energy loss in each elastic scattering event; 4) the
excitation of vibrational and rotational molecular levels during scattering.
The square of the slowing-down length Ar from E = 1.46 ev to E= 0.025 ev does not depend on the initial
energy of the neutrons from the source being used. Therefore if we know T ev for any neutron source, we can
find the square of the slowing-down length.for neutrons of this source 'o 025 eV 1,46 ev Ar to E = 0.025 ev.
For instance, 10.025 ev for fission neutrons slowed down by water is
327
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
r0.025 ev = r1.46 ev Ar = 30.4 2 cm',
since [1] :
r1.46 ev
29.4 1.5 cm2.
In conclusion the authors express their sincere appreciation to Professor I. I. Gurevich for Valuable discussions
of the results of the work, to A. P. Venediktov and B. V. Sokolov, who took part in the preparation of the apparatus
and in the measurements, and to A. V. Telnov, who helped in preparing the apparatus.
LITERATURE CITED
[1] L. M. Barkov, K. N. Mukhin, Atomic Energy. 1956, No. 3, 31.7,?'..
[2] L. M. Barkov,- A. P. Venediktov, K. N. Mukhin , Atomic Energy 1956, No. 3, 329*..,
[3] W. Bothe, 2,-Physik. B120, 437. (1943).
? [4] C. W. Tittle, Nucleonics 8, No. 6, 5 (1951).
.[5] Gamertsfelder, M. Goldhaber, Phys. Rev., 62, 556 (1942).
[6] A. Berthelot, R. Cohen, H. Reel, Comptes rend., 225, 406,(1947).
[7] C. W, .Tittle, Phys. Rev., 80; 756 (1950).
[8] K. E. Larsson, A.rkiv Fysik 2, 47 (1950).
[9] F. J.'Sisk, E: C. Campbell, Bull. Am. Phys. ?Soc., ? 26, No, 1,-58 A (1951).-
[10] G. E. Dardel, N. G. SjOstrand, Phys. Rev., 96, 1245 (1954).
- [11] A...y..Antonoy, A I Isakov, I D. Murin, p. A. Neupokoev, I. M. Frank; F. L. Shapiro, I. V. Shtranikh,
Physical Investigations (Reports of the Soviet Delegation to ?the International Conference on the Peaceful Uses of
Atomic Energy),:Acad.- Sci. USSR Press, 1955, p. 158,
[12] ? F. R. Scott, D. B.:Thornson and W. Wright; Phys. Rev., 95; 583 (1954).
[13].,42.: E. Marshak,:Revs,Mod.?Phys, 19, 185 (1947).
* C. B. Translation pagination.
328
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
SLOWING DOWN OF FISSION NEUTRONS BY. URANIUM ? WATER MEDIA
L. M. Barkov, A. P. Venediktov, K. N. Mukhin
The spatial distribution of resonance neutrons (E = 1.46 ev) which results from
the slowing down of U235 fission neutrons emitted by a "point" source is measured
for three variations of a uranium-water lattice made .of thick slugs (35 mm) of
natural uranium enclosed in cadmium tubes. The absence of anisotropies in the
distribution of slowed-down neutrons is demonstrated, and the values of r determined.
It is of considerable interest to carry out an experimental investigation of the slowing down of fission
neutrons by uranium-water media whose characteristics are similar to those of projected reactors.
In the present work* we measure the spatial density distribution of 1.46 ev neutrons, which results from the
slowing down of fission neutrons emitted by a "point" source in uranium-water lattices, with the aid of an indium
detector. The lattice was constructed of 35 mm diameter blocks of natural uranium enclosed in aluminum
jackets and placed in a tank 140 cm long, 100 cm wide, and 105 cm high. The fission neutron source was a
target-convertor made of uranoso-uranic oxide U235 (4.15 g of uranium). A thermal neutron beam was directed
at the convertor with the aid of a 50 cm long steel tube which was covered with a cadmium jacket together with
the convertor. The distance from the convertor to the tank walls and thewater surface was no less than 50 cm, and
the distance in the direction of the measurement was 90 cm. This made it possible to perform the measurements
up to distances of about 70 cm. In the experiment we used the method of measurement that is described in a
previous article [1]. In order to prevent thermal neutron multiplication, the uranium slugs were enclosed in
tubes made of sheet cadmium 0.6 mm thick. The tubes with the slugs in them were assembled into equilateral
triangular lattices with various spacings of 4.3, 5.0, and 6.0 cm, which provided a variation of the ratio of the
water volume to the uranium volume in the interval from 0,4 to 2.0. The tubes were fixed in position by the use
of three plates with holes drilled through them for the tubes. The upper and lower plates, located at the ends of the
tubes, were made of 6 mm thick Duralumin; the center one, located 20 cm below the convertor, was made of 8 mm
thick plexiglass. The material of the center plate was chosen on the basis of the great similarity between its
slowing-down properties and those of water. The three plates were attached .to each other by a rigid Duralumin
framework. A large window was cut into the top plate to admit the cadmium adapters with the indium foils into
the lattice. In order to maintain the proper spacing between the tops of the tubes that were located within the
window, small sections of the cut-out Duralumin plate were inserted between them..
Figure 1 is a photograph of the lattice, showing the distribution of the tubes (A), and tube (B) through which
the beam enters, the convertor (C), the adapters with the foils (D), the three plates (E1, E2, and E3), and a few
sections (F) for holding the upper ends of the tubes. The tank was loaded with the tubes by use of a winch and
cable with an automatic chuck (G).
In order to verify some questions as to the method used, preliminary measurements were made with a
source that reproduced the fission neutron spectrum. These experiments showed the following;
* The work was performed on the RFT reactor. Preliminary results were presented at the International
Conference on the Peaceful Uses of Atomic Energy in Geneva in 1955.
329
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Fig. 1. General view of the apparatus.
Fr)
-)7 PC64
(2) Ytill
10d-2.(1
(4) water
101 + Measured at a _point
higher than the
source
fp'
to'
?
10"
0 If 10 40 S0 so r.
Fig. 2. Results of measurement of the
resonance neutron distribution density for
uranium-water lattices and for water.
1) In the horizontal plane passing through the source, the density distribution p(r) of slowed down neutrons
is isotropic with respect to the direction from the source.
2) The distribution p(r) in the horizontal plane does not change when a cadmium tube filled with air
(simulating the conditions created by the thermal neutron beam) is placed vertically over the source.
3) The influence of edge effects is not felt as close as 5-10 cm from the boundary, both in the vertical
and in the horizontal directions.
Fundamental measurements were made with the convertor for three different lattices corresponding to the
following values for the ratio of the water volume to the uranium volume: 0.4, 1.0, and 2.0 (spacings of 4.3, 5.0,
and 6.0 cm). The results of the measurements are presented in Fig. 2. The statistical errors for the experimental
points are shown only for the greatest distances r > 45 cm. For distances r < 45 cm, the statistical errors are no
larger than 2 ,"0.
Separate measurements were made for p (r) in the vertical direction, and they showed that within the
experimental accuracy there is no anisotropy in the distribution p(r) in any of the variations of the lattice. Thus
the quantity Texp can be calculated by the usual formula
330
co
f r4 p (r) dr
.t --
6 co
6 f rp (r) dr
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
The results of the calcuiation are presented in Table 1,
TABLE 1
Experimental Values of r for Various Magnitudes of
the Ratio of the Water Volume to the Uranium Volume
Spacing 1 (cm)
4.3
5.0
6.0
Water with-
out uranium
Y = VH20/Vt0t
0.23
0.43
0.61
1
?y = VH20/ Vu
0.4
1.0
2.0
?
,
rexp tcm2)
65 0
47i2
32
'4
29.4 ?.1.5
The values of Tex n presented are somewhat
too high because of the fact that multiplication of
the neutrons incident on the cadmium is not entirely
suppressed.
Values of r corresponding to the absence of
multiplication of these neutrons in the system can be
derived in the following manner.
Let p (r) be the resonance neutron density (for
instance, with E = 1.46 ev) in an arbitrary system
that has neutron multiplication, let p 0(r) be the
resonance neutron density in this system when the
multiplication is suppressed, and let Icc0 be the
multiplication factor for this system. Then neutron
20
0 0.2 0,4 0.8 0,i1 1.0 7,_pii2i2e.
Fig. 3. The dependence of r on the relative amount
of water in the lattice y = VH2ONtot? The solid line
is the experimental value of r: the dotted line is the
value of r corrected for neutron multiplication in the
lattice.
multiplication leads to the following relation
connecting p (r), p0(r), and kcn.
on
911
P (r) Po (r)
kw (r) ? kg. P2 (r) + ? ?
711
where
611
Tr
511
N (r) =
Po ( ? ) po (ri) dri,
411
P2 (r) = Pi (
I) Po (ri) dry
(1)
311
20
NW= Pn-1
r .+ ri I) Po (ri) dr1.
10
Here fpo(r) dr = 1 and f p i(r) dr, . . . fp n(r) dr
are. also normalized to unity. In Expression (1),
kc0p1(r), k203p2(r) . . . represent the distributions
of neutrons of the second and third generations, etc.
If two of the quantities p o(r), p( r) and km
are known, then Equation (1) makes it possible to
find the third. For instance, if p0(r) is known from
experiment, then by calculating pi(r), p2(r) . .
for a known keo, we may determine p (r), or
Fig. 4. Comparison of the data of various works on
the slowing down of neutrons by uranium-water lattices.
0) Direct Oak Ridge experiments (70.022 ev) [3]; A)
Brookhaven experiments; method of artificial "poisoning"
and "enriching"; analysis by the one-group theory (M2)
[3]; 0) Brookhaven experiments; method of "poisoning"
and "enriching"; analysis by age theory (M2) [3]; x)
Westinghouse experiments. Critical.experim ents.
Analysis by the one-group theory (M`) [5]; 0) results of
the present work (rim ev).
331
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
conversely, from the experimental values of p (r)
solved by approximating the experimental curve
tion, e.g. by a sum of Gaussian functions a le
and p 0(r) we can find kce. In practice this problem can be
for p0(r) by a system of functions that is convenient,for integra..
r2
48i , where ai and 8 i are empirically chosen constants. In
this case the integration is completely carried out, and the resulting functions pir), p2(r) are also obtained in the
form of sums of Gaussians with coefficients expressed simply in terms of ai and Ai. It is clear that the necessary
number of terms in Formula (1) is determined by the magnitude of kw. Thus if kea = 0.1-0.2, then the experi-
mental curve for p(r) is sufficiently well reproduced by a formula with only two terms.
P (r) P (r) (r).
(2)
The relation between po(r), p(r), and kcc, also makes it possible to evaluate the differences in the
corresponding values of r if p(r) and km are known, and therefore also p0(r) from (2). Indeed, if p 0(r) is 'given
=
i
in the form of a sum of Gaussians Ea ie , then integration gives
r4p (r) dr
6 S r2p (r) dr
= :k.[ Piai (44)212 +
d?leco:h(Ph+ RL)cit,(4403/2al (4npi)3/2 = +
kc?E(Ph+P)ah (4403/241(41r13031?'
1, it
(3)
In the present work the above method was used to evaluate ro for the system in the case for which there is
no multiplication? neither of thermal neutrons, nor of those that are incident on the cadmium.
In evaluating ro [by Formula (3)] the calculated multiplication factor kep for the neutrons incident on the
cadmium was used, and the value of p0(r) found according to (2) was expressed in the form of a sum of three
Gaussian functions.
The evaluation of the multiplication factor of the system was made according to the formula k =vcoe'e,
where v is the. number of neutrons per fission event, cp is the probability that resonance capture does not occur,
is the multiplication factor for fast neutrons, and e ' is the probability of fission by the neutrons incident on the
cadmium in the process of their slowing down to capture in the cadmium. The value of v was taken from a note
published in Nucleonics [2], that of c from the work of Kouts et al. [3), and that of cp from the results of
measurements by M. B. Egiazorov (private communication).
The value of (). was calculated according to a formula given by Gurevich and Pomeranchuk [4], which the
authors derived for the evaluation of cp and modified for the evaluation of the probability of fission by the neutrons
incident on the cadmium in the process of slowing down. For this the following changes were made in the
formula: the concentration of U238 nuclei was replaced by the concentration of U235 nuclei, the capture cross
section of U238 was replaced by the fission cross section of U2", the term accounting for the contribution from the
strongly blocked resonances was dropped, and a coefficient that accounts for the absorption of neutrons in cadmium
was introduced.
Table 2 presents the values of the multiplication factors kw found in this way for the neutrons incident on
the cadmium. The accuracy of the constants and formulas used allows us to suppose that the error in the
multiplication factor is no greater than 50%. On the same table we present the values of r for systems in which
no multiplication takes place for the neutrons incident on the cadmium. The values of r were derived from the
values of r exp and kco by the method described above.
332
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
TABLE 2
The Multiplication Factor for Neutrons Incident on the
Cadmium, and 7 for the System in the Absence of
Multiplication of Thermal Neutrons and Those Incident
on the Cadmium
Spacing 1 (cm)
4.3
5.0
6.0
Water with-
out uranium
Y = V20/VOt
0.23
0.43
0.61
1
y = VH20tVu
0.4
1.0
2.0
?
kea
0.15
0.11
0.067
0
2
r 1.46 ev (cm )
58 ? 5
44? 3
35? 2
29.4f 1.5
Figure 3 presents the dependence of rexp and
7 on the fraction of the lattice volume that is taken
up by the water y = VH20/Vtot. It should be noted
that the values of 7 found include the effect of
slowing down of the neutrons that arise in the slugs
as a result of On fission by fast neutrons.
Figure 4 presents experimental results of various
works on the measurement of 7 1,46 ev, 10.025 ev,
and the migration area M2.
In making compatisons it should be noted that
+ L2 (L .is
TO.025: M2 = 70.025 ev
M2 is greater than
the diffusion length). For uranium-water lattices,
however, L2 is much less than L2H20, being equal to
L2 ki L0 (1- 0 ) 1-2 cm2.
In addition, in making comparisons we must bear
in mind the fact that the value of 71.46 ev obtained
in the present work is about 1 - 2 cm2 lower than By comparing the values of
70.025 ev? 11.46 ev, 10.025 ev,
and M2 it can be seen that the results of the measurements do not agree well among themselves. Direct measure-
ments of r performed 'at Oak Ridge by a method similar to ours give substantially higher values for 7.
Since the details of the experiment are not described, it is difficult to explain the reason for such a?
divergence. In these experiments no account is taken of the multiplication of the neutrons incident on the
cadmium. As our evaluation has shown, however, this effect is not large. The difference between our results
and the determination of M2 published by Kouts et al. [3] can be partly explained by the fact that in comparable
lattices different amounts of materials that do not slow down neutrons effectively are used.
There is some doubt as to the possibility of determining M2 from critical experiments and from experiments
on the artificial "poisoning" and"enriching" of reactors. In the analysis of these experiments either the one-group
theory, or age theory is used. The validity of these theories in the case of water is doubtful, and the analysis of
the same experimental results leads to -different values of M2.
In conclusion the authors express their gratitude to Professor I. I. Gurevich for discussions of the results of this
work, and to V. K. Makaryin A. I. Maleev, V. I. Baranov, and B. V. Sokolov, who helped in performing the
measurements.
LITERATURE CITED
[1] L. M. Barkov. , K. -N. Mukhin; Atomic Energy, 1956, No. 3, 31 (T. p.
[2] Nucleonics 8, No. 1, 78 (1951).
[3] H. Kouts, G. Price, K. Downes, R. Sher, and Walsh, Experimental Reactors and the Physics of Reactors
(Reports of Foreign Scientists at the International Conference,on the Peaceful Uses of Atomic Energy), State Tech.
Press, 1955,. p..417.
[4] I. I. Gurevich, I. Ya. Pomeranchuk, Construction and Theory of Reactors (Reports of the Soviet
Delegation at the International Conference on.the Peaceful Uses of Atomic Energy), Acad. Sci. USSR Press, 1955, .
p.220.
[5] S. Krasik, and A. Radkowsky,' 'Atomic Power (Reports of Foreign Scientists at the International
Conference on the Peaceful Uses of Atomic Energy), State Energy Press,. 1956, p. 375.
? T. p. = Consultants Bureau Translation pagination.
333
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
RELATIVE Pu239 BREEDING RATIO IN NATURAL URANIUM-ORDINARY
WATER LATTICES
L. V. Komissarov and V. A. Tarabanko
? Measurements were performed of the quotient of breeding ratios
in uranium-water lattices and a uranium-graphite lattice at the start
of conversion. The plutonium breeding ratio in uranium-water lat-
tices for certain lattice spacings is larger than in a uranium-graphite
lattice.
In
An important index of the operation of a nuclear reactor which breeds plutonium is the PU239 breeding
ratio, which is defined as the ratio of the number of Pu239 nuclei (Ng) produced in the reactor to the number
of consumed U235 nuclei (N5):
P - .
N5
At the present time the parameter of uranium-water lattices which are required for the computation of
the breeding ratio are not sufficiently well known. It is therefore important to verify the results of a com-
putation by experimental means.
In the present work we measured the quotient of the breeding ratios in uranium-water lattices and in a
uranium-graphite reactor whose parameters have been well investigated:
Pwater (N9) (N5)(N5) .
graphite
P -(15) graphite (Ng)
graphite - --water
We studied triangular lattices (with 45,55 and 60 Mm spacings) composed of natural uranium and ordinary
water as moderator.
The lattices were composed of avialite tubes 43 x 1.0 mm in diameter containing uranium slugs 35 mm
in diameter and 100 mm long. The slugs were sheathed in 1 mm aluminum. The experiments were performed
on physical uranium-water reactors with a natural uranium zone measuring about 1 m (Fig. 1).
The uranium-graphite reactor had a square lattice with 200 mm spacing. The slugs of natural metallic
uranium with the same dimensions as in the uranium-water lattices were not provided with aluminum sheath-
ing.
Experimental Method
(N9)water
1. Determination of ?
M9).gra. phite
349
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
00000000 0000000 000 00
0 00 0000 00 000 000 000000
0000000000000000000000
00000000000 000000000000
000000000000000000000000
00 000 000000 00 000 000000000
00000000 000000 00 00000 00000
000000000000000000000000000
0000000000000000000000000000
00000000000000000000000000000
000000000000000000000000000000
0000000000000000000000000000000
00000000000000000000000000000000
0000 00 00000 0000000000 000 000000000
0000000 0000 000 0000 000000000000 0000
00000000000000000000000000000000000
000000000000000000000000000000000000
0000000000000900009000000000000000000
00000000000000? 0000000 ?900000000000000
0000000000 0000000 00 000 000 00 000000000000
0 00000 00000000 00 00000000000000000 00000
000000000000000000090000 0000000000000
000000000000000000000$00000000000000
0000000000000000000??? 0000000000000
0000000000000000900900000000000000
000000000000000000000000000000000
00000000000000000000000000000000
0000000000000000000000000000000
000 00 000000000000 00 00000000 000
00000000000000000000000000000
000000000000000000000000000 0
000 00 0000000 000000 0000000 00
00000000000000000000000000
0000000000 00 000000 0000000
000 000 000000 000000000000
00000000000000000000000
0000000000000000000000
000000000000000000000
000 00 0000000 00000 000
Fig. 1. Diagram of a physical uranium-water reactor. $) channels with enriched uranium
slugs; 0 ) channels with natural uranium slugs; ) channels with experimental slug.
It is well known that the production of PU239 in a reactor odours according to the following scheme:
U238 -j--- onl U23?
--->
23,5 m
{3-
Np239 PU239.'
2,3 day
Since the measurements are relative, the determination of the number of Pu239 nuclei produced in a
uranium slug can be reduced to the measurement of the 13 -activity of U239*: In order to determine the relative
quantity of U239 we used disks of natural uranium 35 mm in diameter and 0.1 mm thick placed between the ends
of separated sections of a uranium slug (Fig. 2). The experimental slug was inserted into the lattice and irradiat-
ed with a neutron beam of ? 107 neutrons/ cm2/sec. for 30 min. The irradiation times in the uranium-water
lattices and the uranium-graphite lattice were identical. After irradiation the uranium disk was chemically
cleansed of fission fragments and of the products of natural radioactive decay of uranium by sodium-uranyl-ace-
tate precipitation. The purified uranium was pressed into tablets (of ?300 mg/cm2 density) whose 6-activity
was measured with constant geometry by a Geiger counter. The background of natural uranium decay products
was disregarded since it amounted to only one percent of the measured effect. Measurements of the 13 -activity
of the purified uranium for 1 to 1.5 hours showed that the half-life was 23.5 +0.2 min, which agrees with the
The effect of Np239 consumption during the irradiation period was negligibly small.
350
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Fig. 2. Experimental uranium
slug. 1) Paper disk; 2) uranium
disk; 3) aluminum can; 4)
background paper disk; 5) sections
of uranium slug.
The corrections for this factor,
data of other authors [1]. This indicates that the UZ wwaatsesrufficiently
(
purified of decay and fission products. The ratio, was
(N9)graphite
determined as the ratio of 8-activity of tablets obtained the uranium
disks which were iradiated in the uranium-water and uranium-graphite
lattices.
(N5)g,
2. Determination of (15)w .
The depletion of U235 has two causes; nuclear fission and radiative
(N5)
neutron capture, so that the ratio ---g? can be given in the form
(NOw [Alt. (1+ 21)1
(ivrg+Nr5)G k ) J
1Vr,
(Nr5+1V;)w (1? V )]
(1v5)6
(1v6)w
r
where N5 is the number of U235 nuclei undergoing fission and N5 is the
number of U235 nuclei involved in radiative capture.
The experimental determination of the desired ratio was rendered
difficult by the fact that depletion of 6235 through radiative capture is
?not equal in uranium-water and uranium-graphite lattices.
were estimated from qualitative data on the spectral composition of neutrons in uranium-water and uranium-
graphite lattices as well as the known curves for fission and radiative capture cross sections in U235 as a func-
tion of neutron energy. An estimate of the correction indicated that it is close to unity for all the cases in-
f.
vestfgated. The direct measurement of (N5)g /(N5)w is possible if pure U235 is used as an indicator. However,
we were using natural uranium, for which reason it was necessary to make an experimental correction (Ni) for
fission of U238; consequently, in this case the quantities to be determined are
(Nt -I- NO 6
(Aq + ND w
(I+ 11)
Ng IA/
and
(1+' N-A)
Nt G
Using these quantities the desired ratio can be obtained from the following expression:
The ratio
f f
(N5 + N5,) g
f
(N5 + N5)1,T
(Nt)
(N) w (Nt + ND w
(i iNf)
Nt w
was determined by measuring the a -activity of fission fragments on paper disks of 35 mm
diameter which were inserted between sections of the experimental uranium slug (Fig. 2). The uranium disk for
the measurement of (1?19)w/(N9)g and the paper disk were inserted into the experiment ipetalslug and were irradiated
simultaneously in the neutron field of the lattice. After irradiation the 8 -activity of the fission fragments was
measured with a Geiger counter. The measured activation of the paper disks was 3% of the activity of the col-
lected fission fragments. For each experiment curves were constructed which showed the decay of the activity
with time N = f (t) (Fig. 3).
351
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
(Mt+ Nifi)..c,
( + v./
was determined from the relationship of the areas under the B -decay curves of fission fragments in the same
time interval. The measured 8 -activity of fission fragments will be proportional to the number of fissions with
the same proportionality coefficient if the fragment yields do not change with changes of the neutron spectra
in the lattices. Katkov, using a fission chamber which was described in the paper of Stoliarov et al. [2] found
that the relationship between the number of fissions in a layer of natural uranium and the 13-activity of fission
fragments from natural uranium was constant for all the lattices investigated.
Nff \
The quantity 1-+ 1 +
, which takes
\ N5 N5 ig
into consideration the different contribution of U238 to
the fission of natural uranium in uranium-water and
uranium-graphite lattices, was determined by the method
described in [2].
Number of counts per minute, rel. units
50
109
2R 0 JR J1' 40 4.f .f0 as 58 a 70 t
time after end of irradiation, minutes
Fig. a. Decay of -activity of fission fragments from
natural uranium. )6 -activity of fission fragments
collected on paper disk; ) 8-background of paper
disk.
TABLE 1
Thus the breeding ratio quotient can be calculated
from the following relationship;
Pw (N9)w (Nt+ NO a
P G (N9)0
(1+ 1V1A)
(1+ jN, )
.1Nrl
Lattice
spacing
(cm)
(Ng) Water (Ni6+ Nt) gra.
( 14. NtNr9
\. ' ivt i water
(5'ft )
N , _ grap..I, ite
P water
(No)graphite (Art+ ND water
( 1+ NI3
\- NI J graphite
11 _)P
graphite
(I+
N . . water
4,5
5,5
6,0
1,84?0,04
1,06?0,03
1,02?0,03
1,17?0,02
1,1 ?0,01
1,05?0,01
0,92
0,98
1
1,98?0,1
1,14?0,05
1,07?0,05
Measurements and Comparision with Calculations
All measurements and the calculated correction are shown in Table 1.
The errors shown in Table 1 were calculated from the mean square deviation of the separate measure-
ments.
352
Kunegin and Levina obtained the following formula for the breeding ratio:
(creP)13 iL (I -- cP) liTPsvg
p (ac P)6 1? P?rPP5vE 1?ILTP5v15.
t,cPPL
--psfpoT,
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
where (ci5)8 and (o 5)5 are the macroscopic neutron absorption cross sections for 08 and U235, respectively;"
g is the fast fission f4ctor; cp is the resonance escape probability; vE is the number of secondary fast neutrons
per thermal neutron captured by U235; v is the number of secondary fast neutrons per epithermal neutron
captured by U235; 11,5, pe are the probabilities that an epithermal neutron will be captured by U235 and U233, I
respectively. The values of the quotient of breeding ratios calculated by this formula for the investigated
uranium-water lattices are given in Table 2.
TABLE 2
Lattice spacing (cm)
4.5
5.0
5.5
6.0
P
1.89
1.35
1.16
1.04
8
Fig. 4. Dependence of the quotient of breeding ratios
for uranium-water lattices on the ratio of the volume
P
uranium-water
of water and uranium A values of
uranium-graphite
in a physical uranium-water teactor. The theoretical
curve is solid.
A comparison of the data in Table 1 and Table 2
shows that the experimental values of the quotient
of breeding ratios is in satisfactory agreement with the
calculated values (Fig. 4).
The measurements show that in uranium-water
reactors it is possible to obtain a higher plutonium
breeding ratio than in uranium-graphite reactors.
The authors wish to thank V. I. MostoVoi. M. B.
Egiazarov, V. P. Katkov and G. A. Stolyarov for their
participation in a discussion of the results..
Note: No literature references were given in the
original.
353
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
THE RATIO OF THE AVERAGE FISSION CROSS SECTIONS OF PU239 and U235 IN
URANIUM-WATER LATTICES
V. P. Katkov, Yu. V. Nikolsky and G. A. Stoliarov
The ratio of the average fission cross sections of Pu239 and U235 is determined
for lattices consisting of natural uranium and ordinary water. For comparison the
same ratio was measured for a uranium-graphite reactor. It was found that the
ratio ?af)u/eu for uranium-water lattices with spacings of 45, 50, 55 and 60 mm
and for a uranium-graphite reactor with 200 mm lattice spacing are 2.24, 1.99, 1.88
and 1.79, respectively.
Introduction
The important quantities which determine the physical properties of a,reactor are averaged over the
neutron spectrum of the fission cross section for the elements used as nuclear fuel.
In this article we describe experiments performed for the purpose of determining the ratio of the average*
fission cross sections of Pu239 and U235 in uranium-water lattices and in a uranium-graphite reactor.
The experiments were performed on a physical uranium-water reactor which consisted of two zones. The
central zone contained slugs of enriched uranium, while the peripheral zone contained slugs of natural uranium.
The slugs (of 35 mm diameter) of both natural and enriched uranium in aluminum cladding were placed in
avialite tubes. The tubes were loaded with uranium slugs to a height of 2.5 m. We studied triangular lattices
with spacings of 45, 50, 55 and 60 mm. The measurements were made in the natural uranium zone.
The lattice of the uranium-graphite reactor was body-centered with.200 mm spacing; the slugs of natural
uranium were of the same diameter as above.
Experimental Method
The ratios of the average fission cross sections of Pu239 and U235 can be obtained, as will be shown below,
from the 8 -activity ratio of fission fragments ejected from the layers of plutonium and uranium in the lattice.
Films weighing 0.5 mg were deposited electrolytically on disks of nickel foil 35 mm in diameter and
0.05 mm thick. Quite pure isotopes of plutonium and uranium were used. For safety of operation and in order
to prevent mechanical damage to the films both films were covered with a lacquer thin enough to ensure the
escape of a large fraction of the fission fragments. Special attention was paid to the uniformity of thickness of
the films prepared in this manner.
The fission fragments were collected on disks of filter paper 35 mm in diameter and 0.15 mm thick
placed in contact with the films.
* By average fission cross sections we mean cross sections which have been averaged over the neutron spectrum
and taking into consideration the changes according to the radius of the slugs.
355
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Fig. 1. Position of PU239 and 033 films ma slug of
uranium.
1) Uranium slugs; 2) sealed aluminum holder;
3) U233 film; 4) PU239 filrr,t; 5) paper disk to collect
fragments of Pu239; 6) paper disk to collect fragments
of U233; 7) 15 ji copper foil to shield indicator from
uranium fission fragments; 8) paper "background"
disk.
7000
5000
5000
The plutonium and uranium films and three
paper disks (one was a "background" disk to record
the activation of the paper in the neutron" field) were
placed in a slot in a uranium Slug (Fig. 1), and
irradiated in the lattice for 15 minutes. The 8-activity
of the paper disks was measured successively with a
Geiger-Muller counter. The measurements were begun
20 minutes after irradiation and continued for 30
minutes.
Figure 2 illustrates the decay curves of B -
activity of plutonium and uranium fission fragments and
of the paper "background" disk.
Treatment of the Data
The experimentally measured quantity was the
Apu
ratio ? which was determined as the ratio of the
Au
areas under the .8 -activity decay curves of Pu239 and
U233 fission fragments. The activity is associated with
the number of fissions in the films by the following
relationships:
Npu ilpu.APu,
Nij=r.uAu,
(1)
where Npu and Nu are the numbers of fissioning
nuclei of plutonium and uranium in the films, which
are given by
Fig. 2. Decay curves of 8 - activity of fission
fragments from PU239 and.U233.
N) Activity of fission fragments (pulses/min);
t) time after irradiation (Min.); 1) 8 - activity of
PU233 fission fragments; 2) 8-activity of U233
fission fragments; 3) 8-activity of "background"
disk.
Nu = Wij n (v) Vat (v) dv
(v)v, dv ,
(v)v dv.
(2)
Here Wpu and Wu are the numbers of nuclei of plutonium and uranium in the films; r7-175 is the average density
of neutrons in the slug (with velocity v), which can be obtained from
S2n rn dr
r dr
356
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
R is the slug radius; a Pu and a u are the proportionality coefficients between the number of fissions in the film
and the 8 -activity of the fission fragments.
From (2) the ratio of the average cross sections is obtained:
Using (1) we obtain:
where
atPu WU NPu
-at = WP '
u
? af ' A
_Pu B Pu
A
Go
B =w u aPu
WPu aU
(3)
Thus, in order to determine the cross section ratio it is necessary to know the corresponding value of B.
When this coefficient is independent of the neutron spectrum it is sufficient to obtain it from some known
Apu
is obtained experi-
Au
spectrum.
This can be done by using the relationship -Pu ? BAPU The ratio
Tif a
A ?
"Pu
by calculation. This experiment was performed in a
-t
Apu
combined with the reflector of the uranium-graphite reactor. A
130 cm from the core. The reflector had a thickness of 80 cm. The neutron spectrum in the prism was assumed
to be close to the Maxwellian spectrum at neutron temperature 285?K. --ci,utoti was calculated from the
known curves of plutonium and uranium fission cross sections. It should be noted that for our calculation detailed
knowledge of the neutron spectrum was not required, since in this energy region the plutonium and uranium cross
sections follow the 1/v law approximately. We assumed 0239 = 720 barns and am = 580 barns for neutrons with
2200 m/sec velocity.
The value obtained for B for the neutron spectrum in the graphite prism was 1.70 ? 0.04.
We studied experimentally the dependence of B on the neutron spectrum.
As can be seen from (3) the cross section ratio of plutonium and uranium fissions can be determined by an
ionization chamber which records the number of fissions directly. For this purpose we must know the ratio of
the quantities of fissionable substances in the chambers. But for the calculation of the spectral dependence of
-aPti
B this ratio is not required, since for this purpose it is sufficienz to determine the relative change of
crti
with the lattice spacing, from values obtained by the above-described method with an ionization chamber. From
the identical dependence of these quantities on the lattice spacing we can conclude that B is constant in the
corresponding spectral range.
mentally and the ratio
graphite prism
150 x 150 x 250 cm
was measured at a point
aPu
We compared* the relative changes of
obtained by the methods indicated in lattices with 45
and 60 mm spacings; B is independent of the neutron spectrum in these lattices with an accuracy of 40/0.
* For this purpose we used a double ionization chamber with Pu239 and U235 films. The construction of this
chamber was similar to that described by Stoliarov et al. in the paper "Method of Measuring Fast Neutron Multi-
plication Factors in Uranium-Water Lattices" (Reports at Conference of the Academy of Sciences USSR on the
Peaceful Uses of Atomic Energy, 1955).
357
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
In using experimental results for all of the lattices studied we shall hereafter assume for B the value
1.70 ? 0.04.
Measurements and Results
Apu
As indicated earlier, the quantity was measured in the natural uranium zone. When this zone is of
AU
limited size (? 50-60 cm) it is important to make the measurement in a region which will not be influenced by
the boundary. Then the measured value will characterize the lattice. For this purpose the measurements were
made at several points along the radius of the zone. It appeared that at 20 cm the boundary effect practically
disappears. This was confirmed by control experiments which increased the size of the natural uranium zone to
70-80 cm. The final results given in the table were obtained by averaging the data at three points in a region
where the measurements did not change as the natural uranium zone was enlarged.
Experimental Results*
Oa,
6e,
400
100
Fig.
sections
uranium
VH20
4.5 5,0 45 510 a
Location of
measure-
ments
Lattice
spacing
(mm)
Ap,
-1
?Pu
Au
Graphite
prism
0,80+0,02
1,37+0,02**
Uranium-
graphite
200
1,05+0,02
1,79+0,05
Uranium-
wa ter
lattices
? 60
55
50
45
pertain to
Pu239.
1,05+0,01
1,11+0,01
1,17+0,01
1,32+0,01
a reactor whose
1,79+0,05
1,88+0,06
1,99+0,06
2,24+0,07
lattice
tr.
a567 1000 1460 Z v
040 ,
3. Ratio of average Pu239 and U235 fission cross
as a function of uranium-water lattice spacing.
(71
Pu is the ratio of average and
plutonium
aU
cross sections; a is the lattice spacing in mm;
?
* These values
does not contain
** Calculated value?
is the ratio of the volumes of water and uranium
Vu
in a lattice cell:
The dependence of the ratio of average fission cross sections for Pu239 and U235 on the spacing of uranium-
aPu
water lattices is shown graphically in Figure 3. It follows from the curve that ? in a uranium-water
?1
au
lattice with 45 mm spacing reaches a value of 2.2, i.e., it exceeds by more than one and one half times the
value of the same ratio for a Maxwellian spectrum (Tneun-ons = 285*K) and exceeds by 30% the ratio obtained for?
a uranium-graphite reactor. It should also be noted that the neutron spectrum in all the uranium-water lattices
which we studied except that with 60' mm spacing is harder than the neutron spectrum in the uranium slugs of
the uranium-graphite reactor.
Further studies will enable us to determine in greater detail the characteristics of the neutron spectrum in
uranium-water lattices.
In conclusion the authors wish to thank E. S. Antsiferov for much valuable assistance.
358
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
DECOMPOSITION OF PLUTONIUM OXALATES BY
INTRINSIC ALPHA RADIATION
, ?
V. V. Fomin, R. E. Kartushova and T. I. Rudenko
Decomposition of the oxalates of tri-, tetra- and hexava-
lent plutonium was studied in air and in a vacuum at room tem-
perature and ?80? both under illumination and in darkness, It
was found that the decomposition is caused by alpha radiation
from the plutonium, but in the oxalates of tetra- and hexavalent
plutonium the carbon monoxide which is formed acts as a reduc-
ing agent which transforms the tetravalent plutonium to the tri-
valent form and the hexavalent to the tetravalent. The oxalates
are then transformed into carbonates and, apparently also par-
tially into oxides or an oxyoxalate-carbonate mixture.
In the study of the properties of tetravalent plutonium oxalate it was discovered that protracted storage in
air at room temperature is accompanied by a change of color and loss of weight. The decomposition products
were analyzed by combustion in oxygen as described in reference [1].
The results of .a few analyses are given in Table 1. These show that the oxalate disintegrates during stor-
age, as the relative content of oxalate ions and of water decreases and the amount of plutonium increases, When
the residues formed after storage of tetravalent plutonium oxalate for a few months are treated with acid a gas
is liberated. The plutonium content of the resulting solutions is much larger than would follow from the solu-
bility of Pu(C204)246H20. From these experiments it may be concluded that the oxalate is transformed into a
carbonate during storage.
A more detailed investigation was begun of the transformation of stored plutonium oxalate not only for
tetravalent plutonium but also for trivalent Pu2(C204)3? 91-120 and in part for Pu02(C204)?2H20.
For the purpose of determining the cause of this transformation and the composition of the products a
chemical analysis was made of the transformation, products, the absorption spectra* of the compounds was stud-
ied both in the solid state and in solutions and observations were made of the pressure changes of the gases liber-
ated from oxalates stored in a vacuum. Absorption spectra were taken for oxalates of tri-, tetra- and hexava-
lent plutonium and their decomposition products when stored in air and in a vacuum at room temperature for
various periods of time. The measurements were made at the temperature of liquid nitrogen with a triple prism
glass spectrograph (ISP-51). T,he microphotograms of the absorption spectra were taken with a recording mi-
crophotometer.
In the case of the solid salts a thin layer of the preparation was placed between two watch glasses. A
comparison of the absorption spectra (Fig. 1) of freshly prepared trivalent plutonium oxalate with the products
of its transformation shows that the latter retained the absorption bands which are characteristic of the initial
compound. Some changes of band structure, in particular the disappearance of individual sharp components,
The spectrophotometric studies were made by L, V. Lipis,
409
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Table 1
Change of Composition of Tetravalent Plutonium Oxalate as a Function of Time
of Storage
Storage
period,
(months)
Weighti-
(g)
Content( %
Pu
CO2
2
0.0131
58,8
29.0
12.2
2
0.0212
57,0
28.0
14,8
3
0.0088
61.96
27.0
12.0
3
0.0108
62.31
26.5
13.2
4
0.0084
64.0
24.0
6
0.0166
68.3
17,1
8,8
6
0.0158
69.5
17.4
8,2
17
0,0261
73,5
10.7
5.8
Fig. 1. Microphotograms of a portion of the ab
sorptionspectrum of trivalent plutonium oxalate
and of its decomposition products.
a),For a freshly prepared oxalate of trivalent pluto-
nium; b) for an oxalate stored in air for 2 months;
C) for an oxalate stored in a vacuum for 17 months,
Fig, 2. Microphotograms of a portion of the ab-
sorption spectrum of tetravalent plutonium oxalate
and of its decomposition products.
a)For a freshly prepared oxalate of tetravalent plu-
tonium; b) for an oxalate stored in air for 2 months;
(c) for an oxalate stored in a vacuum for 17 months,
evidently result from the loss of water of crystallization and from the transformation of the oxalate ,into a new
compound. The valence of the plutonium in the resulting compounds does not change.
For tetravalent plutonium oxalate (Fig. 2) the absorption spectra of the decomposition products no longer
agree with the spectra of the freshly prepared specimen;
After limited storage in air and prolonged storage in a vacuum there is a clear reduction of tetravalent
plutonium to the trivalent form. However, a visual study of the absorption spectrum of a specimen which was
kept in air for 17 months showed that the plutonium was mainly in the tetravalent state. This can be explained
by subsequent.oxidation of the previously formed trivalent plutonium.
The microphotograms (Fig. 3) show that when the hexavalent plutonium oxalate is stored a change of
valence also takes place.
This can be determined from the disappearance of the band which is characteristic of hexavalent pluto-
410
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
835mp
nium. In the sulfuric acid solution of a specimen which
had remained in a vacuum for 17 months the absorption
478 mp spectrum revealed the characteristic band of tetravalent
plutonium near 480 my.
708 The decomposition of the oxalates of tri- and
tetravalent plutonium may be due to photochemical
reduction, but this explanation can hardly apply to
trivalent plutonium oxalate. Another cause may be
the action of alpha radiation from the plutonium.
The effect of light was first studied. For this pur-
pose tetravalent plutonium oxalate was put into am-
poules which were kept in darkness for a year. Analysis
showed that the change of composition was similar to
the change observed following storage under light. Con-
sequently, light is not important for the decomposition
of tetravalent plutonium oxalate.
It is easily calculated that if all the energy of
alpha radiation from plutonium is absorbed by the oxa-
late and that the decomposition of a molecule requires
100 ev, the rate of decomposition should be about 0.4%
of the oxalate per day. Therefore it is entirely possible
that the observed decomposition is caused by alpha ra-
diation. In order to verify this, experiments were per-
formed in which tetravalent plutonium oxalate was
kept in evacuated vessels at room temperature and at ?80?C. Samples (containing 50 mg of plutonium) of
freshly prepared oxalate were placed in suspended quartz test tubes which were then put into a glass vessel con-
nected with .a mercury manometer. The air pressure was reduced to 10-3 mm Hg and the vessel was sealed.
The initial position of the mercury was recorded and the pressure was measured periodically. During two weeks
at ?80?C no liberation of gases was observed. After two weeks the vessel containing the oxalate was heated to
room temperature at which it was/maintained for two hours. The pressure then rose to 107 mm; following this
the vessel was again cooled to ?80?C. The pressure fell to 75 mm, which is in satisfactory accord with the
lowered temperature.
nip
883m0
Fig. 3. Microphotograms of a portion
of the
ab-
sorption spectrum of hexavalent plutonium oxalate
and of its decomposition products.
a)For a freshly prepared oxalate of hexavalent plu-
tonium; b) for an oxalate stored in air for 2 months;
C) for a solution of the decomposition products of
hexavalent plutonium oxalate stored in a vacuum
for 17 months.
? 220
4, MO
80
tt.
80 NO 140 920 400 410
Ti,,, e, days
Fig. 4. Changes of pressure during decomposition of
plutonium oxalates over a period of 17 months.
I)Tetravalent plutonium oxalate; II) trivalent plu-
tonium oxalate.
? After a week at ?80?C the pressure had not
changed. The cooling was then discontinued and the
pressure rose to 132 mm at room temperature. The
pressures of 107 and 132 mm correspond to pressures ob-
tained in a parallel experiment at room temperature,
although in the latter case the pressure rose gradually.
Therefore we conclude that the decomposition of an
oxalate at ?80?C proceeds at the same rate as at room
temperature but that the liberated gases are not separa-
ted from the crystals because of a slow diffusion rate.
In our opinion, this experiment shows that the de-
composition of tetravalent plutonium oxalate results
from alpha radiation. As a confirmation we performed
experiments with sodium oxalate. This latter compound was carefully mixed with finely pulverized plutonium
dioxide. The powder was stored for two months. Analysis showed that a pOrtion of the sodium oxalate had
been converted into the carbonate.
These results for the decomposition of the oxalates by alpha rays are confirmed by the experiments of
Bertold Stech [2]. The latter used x-rays to examine sodium oxalate which had been subjected to alpha radi-
ation and concluded also that it had been converted into the carbonate.
In order to obtain an idea of the rate of spontaneous decomposition of the plutonium oxalates experiments
411
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Table 2
Quantity of CO2 and CO Formed by the Decomposi-
tion of Trivalent and Tetravalent Plutonium Oxalates
Table 3
Decomposition Products of Tri- and Tetravalent Plu-
tonium Oxalates
(the samples of oxalates contained 50
utn and were stored for 17 months)
mg of plutoni-
Content (50)
Substance
Initial Compound
Quantity of gas,(g
Pul's
C204
CO0
H20
CO2
CO
Trivalent plutonium
61.6
66.7
trivalent
in evacuated
17.35
15.7
and
7.5
10.8
tetravalent
vessels,
11.2
6.25
as
Trivalent plutonium
oxalate
Tetravalent plutonium
oxalate
0.0030
0.0090
oxalate
Tetravalent plutonium
0.0010
oxalate
0.0013 were performed in which the
plutonium oxalates were kept
described above, but at room temperature. The decomposition was observed for 17 months. The observed pres-
sure changes are shown in Fig. 4, where the measurements made at various temperatures are reduced to 0?C.
The volumes of the vessels were 30 and 27.6 ml; the specimens of the oxalates contained approximately 50 mg
of plutonium.
The curves show the possibility of making'a qualitative comparison of the decomposition rate of plutonium
oxalates of different valences. It is seen from a comparison of the curves that the pressure in the case of the
tetravalent plutonium oxalate increased more rapidly during 2 months than for the trivalent plutonium oxalate.
After 17 months a constant pressure had not been reached but the rate of pressure changed as compared with the
first few days had been considerably slowed down (0.1 mm per day instead of 4 mm). It is difficult to obtain a
numerical value for the rate of decomposition from the data since the pressure is caused not only by the carbon
monoxide and carbon dioxide formed through decomposition but also by water vapor formed through the efflo-
rescence of crystal hydrates and by the'products of their decomposition.
The conversion of an oxalate into a carbonate should lead to the formation of carbon monoxide in an ex-
cited state, which can reduce hexavalent plutonium into the trivalent and tetravalent forms. To verify this as-
sumption we determined the quantity of carbon dioxide and carbon monoxide accumulated in the sealed am-
poules when the oxalate was kept for 17 months. The results are shown in Table 2.
From Table 2 it can be seen that the relative amount of carbon monoxide in the experiments with trivalent
plutonium oxalate is considerably larger than for the tetravalent form.
A number of analyses of the decomposition products were performed; the content of plutonium, C204 and
water were determined by Penfield's combustion method [3] using Berg's buret [4] and potentiometric titrimetry
with potassium permanganate. To measure the carbon dioxide the specimen was dissolved in sulfuric acid _
through which oxygen was blown after being purified of organic impurities, carbon dioxide and moisture. The
oxygen and carbon dioxide formed as a decomposition product entered a flask for titration which was filled with
Ba (OH)2 titrant, where the carbon dioxide combined with the barium carbonate, which was then separated by
filtering. The remainder of baryta water was back-titrated in a 0.1 N solution of Ha.
For the titrimetry it was assumed that all of the plutonium had the same valence; the oxalate content was
determined from the difference between the quantity of permanganate used in titrating the hot solution and the
quantity required to oxidize the plutonium. The average decomposition products obtained after 17 months stor-
age of the oxalates at room temperature are shown in Table 3.
It follows from Table 3 that the composition of these products can be expressed by the formulas
Pu (C204),0. dCO3)13. 4200.221-120
Pu (C204)0, i4 (C 000.6500.2 1-12 00 .25
for initial trivalent and tetravalent plutonium oxalates, respectively.
The oxygen content was determined from the condition of saturation of all plutonium valence bonds.
412
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Thus, a chemical analysis shows that under irradiation the plutonium oxalates are gradually transformed
Into carbonates and later, apparently, into oxides. There is the possibility of formation of mixed oxyoxalates
or oxycarbonates as well as oxalate-carbonates.
When the tetravalent plutonium oxalate is kept in air at room temperature for 17 months the decomposi-
tion is comparatively more complete. In this Instance the specimen contained: Fu02 - 83.6%; C204 or CO3 ?
10.7% (not separated) and H20 - 5.87o.
The reducing activity of the carbon monoxide leads to the result that during the same time interval the
number of decomposing C204 groups in the tetravalent plutonium oxalate is almost twice as large-as in the
trivalent form, per mole of plutonium.
LITERATURE CITED
[1] M. 0. Korshun and N. E. Gelman, New Methods of Elementary Microanal;sis, State Tech. Press 1949.
[2] Bertold Stech, Z. Naturforsch. 2, 175 (1952).
[3] S. L. Penfield, Am. J. Sci, (3) 48, 30 (1894),
[4] A. G. Berg, Factory Lab. 10, 1171 (1948).
413
Declassified and Approved For Release 2013/04/03 : CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
NEWS OF FOREIGN SCIENCE AND TECHNOLOGY
CONSTRUCTION OF RESEARCH REACTORS IN ENGLAND
The atomic energy adminittration of the United Kingdom is buildibg at Harwell two reactors, "Dido" and
"Pluto", intended for broad investigations in physics and technology. Furthermore it is planned to construct a
reactor of the "Pluto" type in Douneray., Inaddition,=England is providing aid to Australia in the building of a
"Dido" type reactor.
Fig. 1. Construction of new reactors for physical investigations at Harwell.
A) Reactor "Didb", designed for general physical investigations B) construction Of a reactor
of the "basin" type, of 100 kw power and with neutron flux 1012 neutrons per cm2/sec ; C)
reactor "Pluto", similar, to the reactor "Dido",, but designed for testing fuel elements.
. .
..Graphite
....Tank. for healiy..Water, . ,keflecto
Ii- . . ...h 200 eM in diameter., a
no K\-\\ti
. ..
..;th 1 Level of If av
:al ---
-
-70 .
- . .
-50 11111;11111;MI&N .jliivall
_
- 3 0
: 1 0 I fittligliMil
- 0 . I Ita
re_ricits__?
5 0
,
-: Fig. 2. .Undisturbed.distribution of thermal neutrons in the
reactor "Dido" (all values in neutrons Per cm2/sec;.).
435
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R00010009p003-2
Fig. 3. 1) Aluminum tank for heavy water; 2) level of heavy water; 3) fuel element; .4) Vertical experimental
channel; 5) guides of Six emergency rods; 6) experimental channel 10.16 x 5.08 cni; 7) biological shield;
8) graphite reflector; 9) experimental channels 36.48 X 20..3? cnii 10) lead heat shield (water dobled); 11) steel
reactor vessel with -boron covering; 12) experimental channel for taking neutrons from the active zone; 13) ex-
perimental channel for taking neutrons from the .graphite reflector ; 14) vertical channel from the graphite
zone; 15) concrete shield; 16) supports of ,reactor; .17) first floor ; 18) thermal column i 19) bracing of reactor
foundation; 20) steel upper lid of thickness 25.4?CM; 21) experimental channel,
' The "Dido" reactor is intended for use in Physical exPeriments, for, example the Measurement of neutron
cross sections, and the study of the parameters of various types Of reactor lattices; besides 'this, the irradiation of
small specimens of materials can be carried out in the reactor: . The moderator andi heat transfer agent of the
reactor is heavy waier,. and the fuel is enriched uranium (about 6:kg:of which 2:5 kg is Um). ,
The active zone of the reactor, with a volume of about 0.3 m3, has the form of a cylinder of height 60
cm and diameter 86 cm. It is composed of lamellar fuel elements,' whose' filler is a mixture of uranium with
. . ,
?
? 436
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
aluminum. The active zone is located in the center of an aluminum tank of height 2 m and diameter 2 m. The
reflector of the reactor is of graphite, 60 cm thick. The reactor and reflector are inclosed in a steel jacket fil-
led with helium.
The nominal power of the reactor is 10 megawatts, with neutron flux 10" neutrons per cm2/sec. The maxi-
mum.possible flux is 5.1014 neutrons per cm2/sec. The starting of the reactor is set for autumn, 1956.
At the end of 1957 the two reactors of the "Pluto" type are to go into operation. The construc,tion and
physical characteristics of these reactors are similar to those of the reactor "Dido', but their purpose is somewhat
different; they are designed for the long-time testing of fuel elements. For this purpose it is planned to install six
or eight testing loops in the reactor "Pluto".
I. S.
LITERATURE CITED
[1] The Journal of the British Nuclear Energy Conference 1, 1, 3 and 35 (1956).
[2] Engineering 180, ,4673, 235 (1955).
A REACTOR WITH AN ORGANIC MODERATOR
According to reports there will be set in operation at the end of this year a reactor with an organic modera-
tor (OMRE) with power 5 to 15 megawatts. The moderator used in the reactor is diphenyl, and the fuel is highly
enriched uranium.
Preliminary experiments have shown the good stability of the organic moderator in the neutron field and
insignificant deposition of precipitates (which had been feared earlier) in the form of a film on the surface of
the fuel elements.
The reactor will probably be built by the Atomic Energy Commission, together with the "North American
Aviatron'Company'in Arco, Idaho, U.S.A. ?
I. S.
LITERATURE CITED
[1] Nucleonics 14, 1, 14 (1956).
[2] Atomic Energy News Letter 15, 2(1956).
ECONOMIC INDICES OF ATOMIC STATIONS
Quite recently there have appeared in the American literature communications on the economic indices
of certain reactors.
Admiral Rickover announced that upon the third loading of an active zone? of the reactor PWR at the Ship-
pingport atomic station the cost of electrical energy will be lowered from 5.2 to 1.5 cents/kw -hr.
* In the reactor PWR it is planned to load successively the assemblies of different active zones.
437
Declassified and Approved For Release 2013/04/03 : CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
At the first loading of an active zone the cost of energy is composed of fixed expenditures (1.5 cents/kw-nx.)
the cost of preparing fuel elements, the cost of zirconium and other construction materials (3.9 cents/kw-in.) and
operating costs (0.3 cent/ kw'-hr).From these 5.7 cents/ kw-hris subtracted' the value of the processed fuel, equal
to 0.5 cent: per kw-hr.
At the second loading of an active zone the total cost of the energy will be reduced to 3.2 cents per kwl.hr
through the lower-cost of fuel, a decrease of operating cost, and the increase of the power from 60 to 90 mega-
watts. At the third loading of an active zone the cost will be lowered to 1.5 cents/kw-hr. For the loading of the
first active zone the working time of the reactor will be equivalent to 8000 hours of operation at full power,
which means, including idle time of the reactor, more than one and one-half years.- The exact working time
with the first active zone is not known, since at first the reactor will be mainly used for experimental purposes.
Its power will be increased gradually: 1st year;. 100/0; 2nd year, 20%; 3rd year, 40%; 4th year, 60%; 5th.year,
either 80% or full power.
The third loading of an active zone in the reactor will be accomplished not ^,arlier than 1965.
The firm "Alco" has released formerly secret data on the cost of electric energy from the portable army--
reactor APPR, which is being constructed in Fort Belvoir (Virginia).
The fuel component of the cost of electrical energy will not exceed 0.95 cent/ kr-hr. This sum includes; -r
0.6 cent/kw-hrcostof fuel calculated at 25 dollars per gram of 05, 0.25 cent/kw-hrfor preparation of fuel ele-
ments, and 0.1 cent/kw-hrfor reprocessing. Operating expenses for an installation of power 2000 kw will amount
to approximately 0.15 cenv*W-hr. Fixed costs will be 0.58 cent-/Rw7hr afloaldlactor 80%.
Thus the cost of electrical energy will come to about 1.68 cents/kW-hr; The cost of energy from an
analogous coal-burning station is 1.4 cents/ kw--hr..
I. S.
LITERATURE CITED
[1] Nucleonics 14, 1, 14 (1956).
A PERFECTED METHOD OF FILM DOSIMETRY
The Oak Ridge Laboratory has developed a film dosimeter with which the following measurements can
be carried out:
a) determination of dose and radiation energy of x-rays;
b) measurement of dose of y -rays;
c) approximate determination of dose of 8 -rays acting alone;
d) approximate determination of dose of mixed 8 - and y -rays.
For this purpose a system of filters is used to simplify the measurements and Make them more exact.
An ideal filter for the measurement of a broad spectrum of x-ray S is one with which the degree of blacken-
ing of the film depends only'on the dose of radiation and not on the energy of the protons. Three types of filter
are described in the report:
1. The filter consisting of the material of the cassette body, conventionally designed "OW" -v-`,"open
window".
2. The "OW" filter plus 1.56 mm of cellulose acetate, called the "plastic" filter.
3. The filter of ? "OW" + 0.25 mm tungsten + 0.5 mm cadmium + 0.5 mm cellulose acetate, called the
"shielding" filter.
438
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
I
"OW" and "plastic"
"shield? "
.
100
200, 300 400
kv (eff)
500 600
700
12
10
as
as
a4
as
2
3
500 1000 1500 2000 2500 .000 3500
m (rJ e:p.)
Fig. 1. Dependence of blackening density of film on Fig. 2. Typical calibration curve for 8- and y -radia -
energy of radiation (dose 200 mr). Upper curve with tions with the different filters.
"OW" and "plastic" filters; lower curve with "shield- 1) yyt of radium, "plastic" filter; 2) y -rays of radium,
ing" filter. "shielding" filter; 3) 8 -rays of uranium, "OW" filter.
24
gu- 'br)22
"..E;
100 200 900 400 500 500
kv (elf):
All three filters are placed side by side in one cas-
sette so that each of them acts separately.
Fig. 1 shows a comparison of the blackening densi-
ties of the film with the "OW" and "shielding" filters
for a 200 mr dose of radiation at various energies. The
curves show that the blackening densities with the "OW"
and "plastic" filters are equal. But for the action of
8 -particles the blackening under the "OW" filter is
greater than under the "plastic" filter, and on this basis
one can determine the dose of 13-radiation in cases of
mixed action of 8- and y -rays from the difference of
the degrees of blackening.
The "OW" filter together with the film wrapper
has thickness 80 mg/ cm2. 8-rays of penetrating power -
Fig. 3. Dependence of the ratio of the blackening
7 mg/cm2 and over are considered harmful to human
densities for "plastic" and "shielding" filters on beings.
the energy of the radiation. Figure 2 shows a typical calibrating curve for various
radiations with the different filters. To determine the
dose of 8 -radiation one needs only to multiply the difference of the readings with the "OW" and "plastic"
filters by 1.5.
The dose of Photons is determined in the following way:
1) from the calibration curve one finds the dose for y -rays with the shielding filter;
2) by subtracting the density of blackening for 8 -rays from that for y -rays with the "plastic" filter one
finds a "correction" density, and then the dose corresponds to 1/10 of the "correction" density. The greater
of the two doses found is taken as the dose of x- or y -rays.
Although the determination of the dose of photons by the method described does not depend on the energy
of the radiation, the filters make possible the determination of the energy of the radiation in effective kilovolts.
This may be needed in case a knowledge of the relative "depth" of the dose is required. The determination of
the energy of the radiation is made from the curve shown in Fig. 3, which gives the ratio of the degrees of blacken-
ing under the "OW" or the "plastic' filter to that under the "shielding" filter as a function of the energy of the
photons.
S. L.
439
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
Declassified and Approved For Release 2013/04/03: CIA-RDP10-02196R000100090003-2
LITERATURE CITED
(1] E. D. Gupton, Radiology 66, 2, 253 (1956).
DOSIMETRY BY MEANS OF GLASS
The majority of existing methods of dosimetry, based on ionization measurements or on the calorimetric
principle, do not provide speed, accuracy, and constancy of readings in the measurement of large doses. The
changes of absorption for the Visible spectrum of most sorts of glass when irradiated with doses of the order of
106-107 r.e.p. make it possible to use glass for the measurement of large doses of .adiation. As a dosimeter
glass has a number of advantages that make its use in this field very attractive. These are: chemical inertness,
insolubility, small size, and durability. But there are a number of factors limiting the use of glass. The most
important are low sensitivity, linearity of readings, and the rapid fading of glass at room temperature.
/0
ZO
0) 01
104 /95
The paper gives data from tests of glass of the
10 r ..1500A 40159,4*
following chemical compositions: a phosphate glass,
2 activated with silver: 50% Al(P03)3, 25% Ba (P03)2,
?