ORIG.RUSSIAN: NEUTRON PHYSICAL CHARACTERISTICS OF U+BE AND U+BEO SYSTEMS

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP88-00904R000100110010-8
• Release Decision: RIPPUB
• Original Classification: U
• Document Page Count: 40
• Document Creation Date: December 22, 2016
• Document Release Date: August 26, 2009
• Sequence Number: 10
• Publication Date: May 1, 1964
• Content Type: STUDY

 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
NEUTRON PHYSICAL CHARACTERISTICS OF U+Be and
Confidential until official release during Conference
U+BeO SYSTEMS
Third United Nations
International Conference
on the Peaceful Uses
of Atomic Energy
A/CONF.25/P/362
USSR
May 1964
Original: RUSSIAN
I.I.Bondarenko, V.P.Garin, R.K.Goncharov,
E.S.Glushkov, O.A.Elovsky, I.I.Zaharkin,
I.V.Istomin, V.G.Kosovsky, A.M.Krutov,
V.A.Kuznetsov, V.I.Lependin, A.I.Leypunsky,
S.S.Lomakin, I.G.Morozov, Uj.A.Nechaev,
V.I.Nosov, N.A.Petushkova, N.N.Ponomarev-Stepnoy,
S.P.Sazonov, O.N.Smirnov, I.E.Somov, E.A.Stumbur,
L.A.Chernov, I.I.Shadura
INTRODUCTION
Beryllium and beryllium oxide due to their good mode-
rating properties and additional neutron multiplication
through the reaction Be9(n, 2n) 'can be used widely in
reactor technology. Various tt+Be, U+BeO systems have been
studied in the USSR at the Kurchatov Atomic Energy Institute
and at the Institute of Physics and Power.
In the paper basic results are given of works perfor-
med during last years.
I. STUDY OF U235 + Be INTERMEDIATE-NEUTRON SYSTEMS
A systematic study of critical paramet8rs and physical
characteristics of intermediate uranium-beryllium systems
with reflector was performed in the stand HO -4 [1] . The
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
critical assemblies were characterized by the equal thicknes-
ses of the reflectors of the same coar )sition (a two-layer
reflector with the inner beryllium layer; sec, Table 1), by
the same porosity (an air volume fraction) of 11.5% and the
aluminium core-volume fraction poor variation.
The critical assemblies forms were hexahedral prisms
assembled from fuel packets filled with a fissionable mate-
rial and a moderator. As rule, the realized critical para-
meters were not minimum for each of the assemblies due to
their non-uniformities which are control rod channels,
safety system rod channels, a core form deviation from re-
gular, etc. To account for these non-uniformities the measu-
rements of the separate core parts and reflector reactivity
coefficients were =ado.
In Table 1 the data on the composition and dimensions
of the core and reflectors are given as are critical para-
meters of some uranium-beryllium intermediate systems studied
in the stand fl -4.
In Fig.1 the urariua-235 critical mass is presented as
a function of the volume with account taken for the experi-
mental corrections for the non-uniformities. The errors of
the critical mass and volume determination with all the
corrections were no more than 1.M.
Particular emphasis have been placed upon the uranium
elements self--shielding effect and its influence on critical
parameters and neutron spectrum. In the critical assembly
III-4-10 (fu 106) the fission density distri-
bution over a uranium-235 plate thickness have been measu-
red. The results of tho neasti3rements presented in Fig.2.i.n-
die3te that the selfsbielding factor given by the relation
4 I =J f (E) d 5i f (~o ) 4O is equal to 0.84. Here
f(o ) is the fission density distribution over the plate
thickness, Jo is the uraniua-235 plate thickness.
Ne ertheless the critical masses of the assemblies
Of -4-10, 110 -4-14 and MM-4-15,, which are with a dif-
ferent degree of fir heterogeneity, were the same
within the experimental errors. The results of these expe-
riments supporti the conclusions of reference [-1] which
362
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
I Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
from moderator blocks and flat fuel elements, which were
arranged in horizontal planes between the moderator blocks.
Caked beryllium oxide blocks and plates with size
100x1OOx5O mm, 50x50x50 mm, 100x50x15 mm and metallic beryl-
lium plates with size 10Ox50x10 min were used as moderators.The
beryllium oxide density in these blocks was 2.8 g /cm3 , the
beryllium plates density was 1.8 g/cm3.
Fuel elements, which were made on base of teflon-4
with 90% enriched uranous-uranic oxide [3I , contained an
average of 1.775 g of uranous-uranic oxide and were with
a square 100x100 mm and a thickness of 0.5 mm.
The effect of neutrons reflected from the walls and
stand constructions was determined experimentally .
All the beryllium oxide moderated assemblies with a
cell pitch of 5 cm had upper and bott)m reflectors with a
thickness of 2.5 cm. The beryllium moderated assemblies
with a cell pitch of 0.5 cm had upper and bottom reflectors
with a thickness of 0.5 cm. In the experiments the efficien-
cy of the reflectors was found to be equal to their geomet-
rical thickness with such reflector Sizes.
The experimental data obtained on critical assemblies
were used in calculations performed numerically by the niul-
tigroup method (41 on a digital computer and using the
relation
2
K~ e
Koff = --- -_II L
27-2---- (1)
The critical assembly parameters and calculational
results are presented in Table II.
Due to absorpticn in the reaction Be9 (4 , n ) of
(r -rays from long-life fission products photrneutrons,
which should be taken into account in the transient proces-
ses analysis, are producted apart from the delayed neutrons
in systems containing beryllium.
Using the data on the delayed neutron and photoneutron
yields reported in reference [5] , the experiments consi-
sted of measurements of reactivity, delayed neutron and
photoneutron efficiencies were processed, calculations of
the errors due to the asymptotic period measurements were
performed .
762
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
The deviation of the measured positive period from
asymptotic d was determined for different times of
reactor operation at constant power level. Calculational re-
suits are presented in Fig.5. For the beryllium oxide mo-
derated assembly the deviation of the measured period from
asymptotic was also determined experimentally (see, Fig.6).
For the beryllium oxide moderated assembly the quantity
which was assumed the same for all the groups of the
delayed neutrons and photoneutrons, was determined by the
method of replacement of fuel by an absorber.
In these experiments plates especially fabricated from
teflon-4 with boron filling were use4 as a~sorbin,-; elements.
The experimental procedure consisted in the replacement
of fuel elements by absorbing elements and in the reactivity
measurement in mm of the control rod length.
,It proved Y to be equal to 1.15 for the beryllium
oxide moderated assembly studied; the experimental error
was 4%.
The calculational ~ value proved to be equal to
in assemblies experiments were performed in which the
critical assembly reactivity variation with a small change
in the assembly dimensions was determined.i~xperimental
data processing was conducted using; the relation.
H [2Y z (`C + L2/1 + B2L2113 (ah,/kcffa 1
where H = h + X , h is the core size, X is the extrapola-
ted addition or the effective thickness of a reflector in
case of an assembly with a reflector.
Using the beryllium oxide and beryllium diffusion para-
meters reported in this paper the values werte deter-
mined from these experiments and from the critical condition
Kco was determined.
The data obtained are presented in Table III.
The neutron spatial distribution measurement was used
to determine the neutron physical parameters and subcri t ica-
lity degree of the systems studied.
In the case of a system, form of which is a rectangular
parallelepiped, the ratio of the neutron flux in an assembly
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
with no uranium to the neutron flux in a subcritical assemb-
ly with uranium can be written as
_ L
~ S 2 urw ~ u? x ~ wJr~
vw uvw ~~+~ uvw L CoS CL cos SOS
u C
WT
uvw 2: c (1 + l31Uv v L )11- kuvw e3 f ~s Q'3 G)S b 9 C)s
This relation, which can be considerably simplified if
to neglect by the harmonics of order higher than third and
to check such an experimental geometry that the third harmo-
nic becomes zero, was used to determine the beryllium oxide
moderated critical assembly subcriticality degree.
The neutron distribution was measured in experiments
with indium foils.
The measured thermal neutron distribution in assemblies
with no fissionable material and with a neutron source in
the centre was used to determine the age value T . For be-
ryllium oxide T of the Po + Be source neutrons proved
to be equal to 128 ? 13 cm2.
To study the neutron physical characteristics of a beryl-
lium moderated assembly with two fuel elements in a layer
the pulse-source method was also used. In the experiments
a neutron generator of neutrons with energy of 14 MeV was
applied. The damping constr.nt dependence on the geometrical
parameter was established from the experiment. The treatment
of the dependence was performed by the least square method.
The following parameters were obtained:
Ka,= 2.23?0.05; L2= 20.7?1.5cm ;
i00 =
(1.74 ? 0.18) . 10-4 ce)c 1
KBe =
1.12 ? 0.05;
j, =
(6.68 ? 1.08).10-3.
III. STUDY OF U-Be LATTICES WITH NATURAL
AND LOW ENRICHED URANIUM
The experiments were performed in a critical assembly
which was a cylinder of metallic beryllium (mean density
t= 1.75 g/cm3) with a diameter of 116 cm and a height of
100 cm, surrounded with an all-sided graphite reflector of
60 cm thickness. In the critical assembly core there were
- 6 -
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
89 vertical holes of 45 mm diameter located in chess-board
order, which permitted to form lattices with pitches of
11.3 cm or 16.Ocm. From the core centre a horizontal hole
was extracted of 40 mm diameter for the measurement of
the neutron spectrum over the cell.
As fuel natural and 2 enriched uranium blocks assemb-
led in worker channels were used. The blocks were uniform in
their dimensions and configuration and differed only in enrich
went and U235 content per block unit length.
The core dimensions permitted criticality to be achieved
with 2% enriched fuel for both indicated pitches. To study
the multiplication factors of natural uranium lattices
two-region systems with the natural uranium central region
were assembled . Criticality was achieved with the external
region, which was assembled from 2% enriched channels. In
Table IV the critical dimensions are given for 27o enriched
fuel.
For four lattices there were measured the infinite neut-
ron multiplication factors koa and the thermal neutron
spectra over the cell.
The experimental measurement of kc-
procedure.
Accordingly to [6]
was based on a method reported in reference [63 with
rather modified measurement technique and results processing
v +(Ug -1) n2f/nsf
.~ ns%ns~ + n3%~~+
f 5
4) ~
~V
~n
8-
8
=1 +-
s-----
v (8)
R f1
ns /nst +n3/rts~ +a %r15 r9)
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
tl s n5 + + n $ T'~/ n'
ns%n + ~8 r jn
- number of fidsion neutron absorptions
in uranium-235 and uranium-238
relatively;
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
relatively total number of neutron radia-
tive absorptions, number of radiative
absorptions in the resonance region and
number of radiative absorptions due to
-1- cross-section in uranium-238
V 91 nR =n8 + ng
number of neutron absorptions in moderator
and structural materials.
The number of neutron absorptions in lattice components
was determined from iueasuremen6 of relavitve activities of
detectors, which were exposed in the central cell of stu-
died lattices. The detectors were 35 mm - diameter foils
of uranium-235 (9097 enriched), depleted uranium (3.103%
of uranium-235), natural uranium, thorium, vanadium and
c opper.
A packet of several detectors, 121in ( 1.5 mm) uranium
washers of a lattice studied and protective aluminium foils
was assembled in a special experimental channel. (Fig.7)
and was placed in the central cell for exposure.
The absorptions number in beryllium ( f7j ) was deter-
mined from the measurements of the neutron density distri-
bution over time cell (Fig.8) with copper and vanadium detec-
tors. Exposure of all the detectors was performed both with
no cadmium and with cadmium jackets. The correction for
"eating away" of the resonance and fast neutron flux due
to cadmium was determined experimentally.
The detectors activity measurement was performed with
a scintillation counter and a 128-channel amplitude ana-
lyzer. All the measurements of the absorptions in lattice
components were made relatively to the number of fission
neutron absorptions in uranium-235 (M- ) .
Using the experimental data and formulae (4+8) Ka., ,
if , B , V e ff and P (breading coe.,xfiaLent) were calcu-
lated of the lattices studied. The calculational results
are given in Table V.
362
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
In the multiplication parameters calculations the fol-
lowing constants values were assumed
~s = 2.43 -0.02
J~ = 2.9 ? 0.1
~5 = (582 ? 4).10-24 cm 2
cm2
C~7g = (2-75+-0.04).10 24
By Urtseva V.I. the parameters experimentally determi-
ned as functions of a lattice pitch have been calculated in
the two-group diffusion approximation. For comparison the
calculational and experimental K co values are presented
in Fig.9.
In the uranium-beryllium systems study the thermal
neutron spectra were also measured in several 2% enriched
and natural uranium lattices with a mechanical monochroma-
tor [91 .
The thermal neturon spectra obtained on a cell boundary
for two lattices are given in Fig.10 . The experimental
points were processed by the method of least squares with
a crude description of the spectrum near its maximum as
the maxwellian distribution. It is appeared that only the
spectrum over the maximum can be fitted to the iaxwellian
distribution that conforms to the calculations by Nelkin and
Cohen [10] .
To compare the experimental data with the theoretical
results by Smelov V.V. anci Ilyasova G.A. the 15-group cal-
culation have been made of the thermal neutron space-energy
distribution over the cells studied in a P3 - approximation
using a model to account for the crystal bindings in the
thermalization region.
IV. STUDY OF BERYLLUM REFLECTOR EFFECT IN FAST
NEUTRON REACTOR
A decreasing of beryllium fraction and an increasing
of fuel enrichment in the reactor core leads to the
neutron spectrum hardening.
As a limiting case, in this sense, a reactor can be
considered in which the core contains only fuel of maximum
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
enrichment, and moderator (beryllium) is present only in a
reflector.
Such a system can be identified as a "fast reactor with
a moderating reflector" x).
In w958-1961 two physical beryllium reflected fast
reactors, BP-4 and BP-6, were studied. The rqactor shapes
were right hexagonal prisms. The cores were assembled from
almost-close-packed rod-type fuel elements (plutonium ele-
ments in EP-4 and 90% enriched uranium elements in EP-6).
The beryllium reflectors can be replaced in the thick,.
ness range from 4 cm to 15 cm.
For beryllium reflector thickness more than 5-6 cm
the spatial distributions of the main reaction (U235(n,f)
in BP-6 and Pu239 (n,f) in R-4) density differ essential-
ly from the distributions in common fast reactors with non-
moderating reflectors.
For this fission density flattening occurs over the core
radius due to intermediate energy neutrons penetrating out
of a reflector.
Near the core boundary a sharp fission density jump
forms,for the most part, due to subcadmium energy neutrons.
Whereas the U238 fission density behaviour is practically
independent upon the reflector thickness with asymptotic
character held (see, Fig.11).
The main iso'ope fission density jump on the core boun-
dary (a difference between the true and asymptotic fission
densities) increases with the increased reflector thickness.
With the reflector thickness 4 = 546 cm. the jump disap-
pears, i.e., a more thin beryllium reflector does not already
make cardinal changes in the fission density distribution.
The central core' part for all the LP-4 and EP-6 va-
riants maintained a hard neutron spectrum. The typical
ej (U235)/4(U238) value was ---7 and -9 in the EP-.4
centre and in the ?P-6 centre respectively.
x) Similar fast-thermal systems have been studied until
recently by a number of authors E11 ,. 12, 13]
- I0 -
Approved For Release 2009/08/26: CIA-RDP88-00904R00010011001
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
independent cross-section i a . i.e. . of fihc 1/F. m
slowing-down spectrum in an infinite medium with the energy-
down neutrons spectrum form can be obtained similar to the
there is a large neutron leakage in slowing-down. The slowing-
can not be considered as the pure Fermi spectrum, because
The slowing-down neutrons spectra in thin reflectors
spectrum of neutrons returned by it change.
The thickness of the reflector varying, its albedo and
of 12-15 cm.
reflector thickness of 8-10 cm and 4t 57, with the thickness
ween the core and the reflector. It equals 1 + 2% with the
from change of Keff due to a cadmium shield inserted bet-
8-12 cm). The thermal-neutron fission fraction was estimated
E >1.4 Mev was 20-30', (with the reflector thickness of
cated that the fission fraction due to neutrons with energies
Intergra-ting of the fission density over the core indi-
line which corresponds to 1/Em , and also the spectrum hys-
togram obtained from the multigroup calculationx) are shown,
The simulation of the spectrum in a moderating reflector
with account taken for a possible distinction from the Fermi
{p2
The Ya, role in this case is played by 2
where B2 is the geometrical parameter of a reflector with
the thickness d
In the age model in = 1 l2s B2/3 f (4 )
When the reflector thickness decreases from 15 cm to
8 cm, the value of m decreases from 0.9 to 0.7 . With the
reflector thickness, Z. 4 cm value of m L 0 , i.e. the
flux spectrum does not increase with decreasing of energy ,
but decreases that is typical for fast reactors. Inter-
pretation carried out shows that a limit thickness 4 -4 cm.
(m = 0) exists which as if separates the common fast reactors
from the class of the "fast reactors with a moderating ref-
lector". The 1/Em spectrum form was confirmed by S.A.Kati-
shev's measurements of the resonance detectors activation
in reflectors.
In Fig.12 the results of the measurements, the straight
form (see,[15J ).
x) The calculations were performed by E.I.Lyashenko under
the guidance of G. I . Marc huk .
362
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
spectrum permits, for example, to eliminate the discrepancy of the
experimental and calculational Rcd values reported in reference [IIi,
In the thin reflectors considered the incomplete neutron therma-
lization effect becomes also apparent which results in the anomalously
high ratio of the subcadmium Uf (Pu239) / .t (BI0) .
The prompt neutron mean life time "1" in the systems considered is
essentially less than in thermal reactors, but more than in fast reactors
The measurements made by the Rossi-p( and boron poisoning
methodsx) (with A eff - 0.007) gave for the reactor 5P-6 the values
which increase with the increased reflector thckaess: with 8 cm ;
1 = 10 6 sec; with Q = 12 cm; 1 = 5.I0-6 sec.
The performed measurements of disturbances allowed to verify an
applicability to moderating reflectored fast reactors of the "degree
law" which characterizes through the exponent n(x) the effect of the
each diluent "x" on critical mass (see, [14] ). The measurements shows
that the values n(x) decrease with the increased reflector thickness and
can even be negative.
In 5P-6 with a = 10 cm: n(void) = 1.44 ; n (U238) = 0.95;
n(Be) = 0.84 and n(H20) = 0.30. And with a = 14 cm: n(void) =0.83;
n(U238) = 0.4; n(Be) = 0.29; n(H20) = -0.20
I. A.M.JlevinyHCKMM LI Ap. "31ccnepl1MeHTaJILHble HccJIeAotiaHZH HeKOTOpbIX cI143K-
uecxJx oco(eHH0cTez npOMemyTOTHbLx peaKTOpoB c 6epYIJIJIYIeBbIM 3aMeAJIYITeneae
Proceedings of the Conference on Fast and Intermediate Reactors
Physics. Vienna, 1962.
2. B.A.Iy3He1~oB Id Ap."rMApubi meTaJIJIOB Kax 3aMeA7IMTeJI1I HeYITpOHOB
B peaxTopax". Proceedings of the Conference on Fast and
Intermediate Reactors Physics. Vienna, 1962.
3. H.H. HoxoMapeB-CTenIHOVI, C.C. JIoMaHHH, 10.r. AeraJibueB,
ATOMHax MeprvxH, 15. 259 (1963).
4. P.M.Mapw. "MeTOAEI pactieTa HAePHb1X peaITOpoB". ATOMK3AaT,
MocxBa, 1961.
5. G.R.Keepin, Nucleonics, 20 (1962) 151.
The measurements were performed by D.M.Shvetsov and V.H.
Manohin.
362
- 12 -
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
6. C.M.Te H5epr H AP. BblropaH1Ie roprOtiero B BOAo-BOARHbIX 3xepreTUINcHVIx
peaHTOpax PI 31CCnepM HTbl c ypaH-BOAHOM pemeTHO t.
Proceedings of the- Second International Conference on the Peaceful
Uses of Atomic Energy, Geneva, 2145 (1958)
7. A.K.Kpac1H UI Ap. 13yueHHe c1li13vt,IecKr1x XapaHTepKCTIM peaKTopa c
6epMAJIMBbIM 3aMeAJIMTeneM.
Proceedings of the Second International Conference on the
Peaceful Uses of Atomic Energy, Geneva, 2146 (1958)
8. UI.c.}KefepyH, 1.II.CaAMHOB, B.A.rapa6aHEHO, A.A.4epIIbIweB.
ATOMHa1 3Hepr1R", 15, 485 (I962_).
9. A.D.CexueHxoB, (P.M.Ky3He1 B. "ATOMH85 3HeprMR", 5, 124
(1958)
10. M.C.HenIHVIH, E.P.Koy3H. "HegTpoxxa 1 (1I3IQtta" (c6opH nc), 654,
M. ATOM1I3AaT, 1959.
II. R.D.Smith and J.E. Sanders "Experimental Work with Zero Energy
Fast Reactors". Proceedings of the Second Conference on the
Peaceful Uses of Atomic Energy, Geneva, 39 (1958)
12. R.Avery et al. "Coupled Fast-thermal Power Breeder Critical
Experiment. Proceedings of the Second International, Conference
on the Peaceful Uses of Atomic Energy, Geneva, 2160 (1958)
13. H.H.Hummel et al. "Experimental and Theoretical Studies of the
Coupled Fast-Thermal System ZPR-V". Proceedings of the Second
International Conference on the Peaceful Uses of Atomic Energy,
Geneva, 599 (1958).
14. G.E,Hansen et al. Nucl.Sci. and Engng. 8, 543 (1960)
15. Weinberg A., Wigner E. The Physical Theory of Neutron Chain
Reactors. University of Chicago Press, 1958.
- 13 -
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 ?d
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Critical Parameters of the HO-4 Assemblies
U)
0
U) ;,
h O U
U
c 4-I
n - O
O c: ~, u~~ p v Ed
v rl
r-i
N C ?rl O 5 ij
4-3
U /
U
~ cU CU
O U
C "
PQ ?ri ?rr-1 f 1
- v O
a) U)
U2 -H~4 43) CO 41 4-,;--, 1 - r- Fa
421 ~j O a) r-{ U
U U O U U .r{
O O O p H 94 ~J 4-D p ?rl C r-j 01 IN C,
110 -4-09 0 0 0 13.9 9.2
HO -4-07 1.76 0.5 0.5 13.9 13.9
HO -4-13 12.8 0.5 0.12 13.9 9.2
64%
2.0
(36%)
110-4--08 53.6 0.5 0.03 13.9 13.9
IIID-4-10 106 1.0 0.03 13.9 9.2
III -4-.11 157.9 1.5 0.03 13.9 9.2
110-4712 209 2.0 0, 03 13.9 9.2
110 -4-14 106 2.0 0.06 13.9 9.2
110 -4-15 106 3.0 0.09 13.9 9.2
20.81 0 64.4
25.22 27 42.7
25.29 59.2 13.5
18.15 70.2 4.0
14.26 72.4 2.0
11.84 73.1 1.3
10.14 73.5 0.9
14.15 72.4 2.0
14.62 72.4 2.0
13.9 28.9 31.3 28.34 0.4
13.9 31.2 41.8 42.76 0.2
13.9 39.2 41.8 53.69 0.15
13.9 43.0 41.8 58.89 0.10
13.9 30.9 41.80 42.41 0.2
13.9 32.0 41.843.83 0.2
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
-----------------------------
 11111000 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Beryllium oxide and beryllium moderated critical assemblies
Moderator Number of U-235
fuel ele- mass
ments in 191
a layer
Table II
---------------------------------------------------------------
cell pitch Atomic B2.102 K eff cal- Keff multi-
CcmJ ratio Ccm-2] culation
using group
BeO/U-235 experi- calculations
mental
parameters
---------------------------------------------------------------------------------------------
Beryllium
2803
5
8699
0
274?0
005
1
040
1X)
.
.
.
4
1.01623
3060
5
4904
0,376}0.005
1.01423
0.99028
3409
5
3272
0.438}0.005
0.99872
0.96684
4225
5
2464
0.478+0.005
0.99178
0.96739
9
4828
5
1974
0.488+0.005
0.98878
0.97191
4
3528
110
4907
0.3 57?0.005
1.00642
0.98028
6
4232
10
3262
0.414?0.005
0.98474
0.97070
8
5146
10
2464
0.455?0.005
0.96606
0.96468
1U
5'49
10
1955
0.463?0.005
0.96977
0.97808
3292
1
3481
O.b54?u.008
0.990u2
0.98206
5790
1
1745
0.677?0.008
0.99785
0.9537
8344
1
1165
0.666?0.008
0.99574
0.98578
x) An assembly with non-uniform fuel distribution which was not taken into account
in the calculations.
E Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110
Table
Experimental data for assemblies fro
beryllium oxide and beryllium
---------------------------------------------------
Moderator Number of fuel
elements in a
layer
Beryllium
1
2.24?0.06
Beryllium
Beryllium
2
2.28?0.06
1
oxide
3
2.07?0.06
1
Table
Critical, charges and critical sizes for
2% enricned fuel
Lattice Core height Care radius U-23
pitch [cm] [cm] char
Cc MI [kg
16.0 100 60.5 4.50
11.3 100 37.2 3.40
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Table V
Multiplication parameuers in uranium-beryllium systems
1.06 ? 0.01
1.31 ? 0.06
0.76 ? 0.03
0.92 ? 0.01
1.05 } 0.03
0.98 } 0.03
0.97
1.08
1.36
1.50
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
to Z. 3Ntptul E, .r!
Fig.4. The neutron flux energy distribution in the 110 -4
assembly measured in the central and peripheral
regions. The curves are normalized to power.
0- measurements in the central region measurements
? - measurements on the beryllium reflector boundaryr
1 - flow 0 (E) , relative unity;2 - energy E, eV.
:
;
:t::E
41 Fig. 5
Fig.5. The wait time before the period measurement follo-
wing the reactivity jumpy as function of the reacti-
vity value for various O .
A(d) beryllium moderated critical assembly,
T 1000 sec.
_ = 0.1.10-21 1 = 10-4 sec.
t - time following the reactivity jump, sec.
- reactivity in per cents.
_19-
Approved For Release 2009/08/26: CIA-RDP88-00904R000100
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
N
Fig.6. The deviation of the measured period from the asympto-
tic period as a function of time following the reacti-
vity jump,.
A(d) beryllium oxide moderated critical assembly,
T = 1000 sec, jo = 0.1.10-2 , 1 = 10-4 eec .
I - calculation with no account for photoneutron
2 - calculation with photoneutrons taken into account.
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Fig-7.
Design of a work fuel element (a) and an experimental
block (b).
1 - uranium blocks; 2 - Al block can; 3 - aluminium
work channel tube; 4 - upper and lower parts of an
experimental uranium block; 5 - uranium washer of a
1.5 mm thickness; 6 - protective foil of a 0.03 mm
thickness; 7 - Al container; 8 - Al plug of the
experimental channel upper half; 9;,;- fastener to at-
tach the experimental block upper part to the experi-
mental channel upper part; 10 - detector to be expo-
sed; 11 - cadmium sneath; 12 - aluminium ring
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
r
A7 CM
v
!O c
Fi;.8. The thermal neutron density distribution over a cell:
a and b - 2% enriched uranium lattices with a pitch
of 11.3 cm and 16,0 cm relatively;
c and d - natural uranium lattices with a pitch of
11.3 czu and 16.0 cm relatively;
1 - uranium boundary; 2 - Be moderator boundary;
3 - Al block can boundary; 4 - equivalent cell
boundary (1 - cm).
a
Fig. 9. The uranium-beryllium lattice multiplication coefficient
as a function o.C the lattice pitch:
I - calculational curve for natural uranium;
2 - calculational curve for 2% enriched uranium .
(1 - natural (1)
362 - 21 -
Approved For
*-fill
F n
l
l
/Z e
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
55
V
Fig.1Q.The thermal neutron flux spectra on the cell boundary
(0 - cm).
a - 2% enriched uranium lattice (or a 16.0 cm pitch) ;
b - natural uranium lattice (of a 11.j cm pitch).
i - calculational spectrum for the model with account
taken for the crystal effects; 2 - neutron flux se-
wellian distribution curve.
Pig 11. The relative fission density distribution over the
cell radius ( e is the reflector thir-kne in on)
Pig.12. The fission densi inn ?C~ ~, th ,~_ sec
as a t utic of the reflector thicIresg.
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Third United Nations
International Conference
on the Peaceful Uses
of Atomic Energy
A/CONF.28/P/362a
USSR
May 19 64
Original t RUSSIAN
Confidential until official release during Conference
I.B'.Zhezherun, A.K.Kraeln, tl.l.l'lindov, I.P.13adikov, V.A.Terabanko,
A.A.Chornyalrgv '
Beryllium and boryllia are of .int'rent l.n the nuclear engineering an construction mitte-
r?sla of roactore of various types due to the'.r good moderative proprrtlee, weak thermal neut-
ron absorption and fast neutron iou.ltiplication in the reaction 110q(n,,.'n), Physical proportien
of beryllium and beryllia determining dlffualonr moderation, scattering and multiplication of
neutrons have been etudlod for a few years In I:he Kurchatov Institute of Atomic Energy (1AB).
The main results are briefly presented in the report.
In ail.lttion, the feet neutron multiplication data obtained from the analysis of the criti-
cal assemblies in the Institute of hnergotlcn of Academy of Science BSEIB arc preacnted.
I. F.Zhozherun, I.P.Uadikov, V.A.Tnrabenko,A.A.Chern.ynhev
(The Kurchotov Institute of Atomic Raerpy)
1. Effect of beryll..u microstructuro and temperature on thermal neutron
scattering aronts-section
The microcrystallyne structure can essentially influence on the moderative and acott'r t,
properties of boryllia due to the effect of size and orientation of crystal grains on the
scattering cross-section 65 . It 1s known that with the increasing of the groin (OS
decreases due to the extinction effect; their advantageous orientation (texture) which can .el-
pear in preening or pressing out of BeO articles can cause the anisotropy of 6S . To make cle-
ar how much those effects are to be taken into account when cnluAating the reactors of BB"0
moderator, the total cross-section dt (6t t cf for BeO) was mess fired with the chopper for
four samples with various grain size (see Fig.I). Uauple I consisted of the plates I.2z7x'1 ria
produced by pressing out, sample 2-of the rods 3x4x23 ma produced by pressing, sample 3 on le
of the disco 30 mm in diameter, 3 mm thick produced by pressing out and sample 4- of the discs
38 mm in diameter, 3 mm thick produced by cold preening. The microstructure enaly min of the sa-
mples, carried out by Yu.G.Digaltsev and V.I.Kushakov, showed the all samples are cloned peeked
crystalline of irregular polyl}edron form. The grain sizes are shown in Fig.1, the moon size was
determined as a square root of an average area of a grain in a microseotion photograph. In
different measurements the sample thickness varied between 3.2 g/em2 and 5.6 g/om2.
Under given experimental conditions the energy resolution arse between 4 and 10% in the
range 0.03-0.1 ev. To observe the anosotropy of 6., two series of measurements have been done
with samples I and 2, the neutron beam being fallen both in parallel and normally to the d i-
rectione of pressing or pressing out (Fig.2).
It is seen from these figures that at small energy (0.005-0.2 ev) the difference in ash
is 40%, for the samples with the grain nizns 8rt. and 29_~L i it may cause 10 .-char-rc of the 'o "-
rage spectrum scattering length ~.S or the square diffusion length Li (see Table 1). The ex-
tinction effeot does not practically influence on dS` if the grain also 10}t
The proeeRee of texture in the hexagonal B9O lattice enema to be observed in the beet -
from neutron wave reflection on the planes being normal or parallel to the -axi3 of the crye-
pproved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
tel elementary eel I3 I.e. from the rof.loetiona to whlcl+ tilt, first. two indexes ur the loot one
are equal to zero. The antnotropy of 64 in obuerved from the reflection on the pliwo (110)
at F = 0.012 ov (nr?o inserts on Fig..') as renchtng of the 1-nrtlr-1 or 'J;x of the measured
crone-aectiono. The values of the cross-neetionn #5 140 "rid , uhow, the arlernntnvstiou i orien-
tation of the cr?ystn.l. g;rainn by c-axin of the c.Irnm,ntnry cell in direction of ire's iing out and
normally to the preening dtr+iction, if the nnut.run bewn Is sorest or pavellel to the prenuing
out or prensing directions. For other refiuctluna the differronce in 46j and (b# , if It 19,
does notZexceed the measurement: er+r'r?n. 'Thou, the enilietropy of in tunell and its inriuen-
oe on b may be neglected.
In the range 1-10 ev where atom binding in the crystal lnttlce doPn not effect on scatte-
ring the measured value of 6, in connt,int rend equals 1(?1). 15 born for n 1 1 the anmploe.
Tho temperature depentionce of Get it; of, tmportonco for the colr:ultitton of the tomporatu-
ro reactivity coefficient. In Fig.3 the ins ?ru;uremeut. results of c6,L ere shown for the sample
of advantageous grain sizes 10-'Opt at 'I' "Jot 11,K); '13001 1500'K, In Flg.4 the tomprrruttuo
dependence of (6, for some energies Lu plotted an wrrl l nn that of jt and (6f 1)_f uvernged
over the Maxwellian spectrum (tit the neul;ran trmperatearc TH : T + 1(U"K) and obtnin+rd from
Fig-3.
The measured results reveal that with the sample hentinri 1) tit,) pt-tiI.fon of Ili-n;!; mnxima
are shifted towards lose energies In accordance with the temperature expnnnienl .') the voluo
of the maxima decreases; 3) the value of (e t tncrnanerr. the ernetor it. ev11) he Leo lo -it, tine
neutron energy E is. At F. 7 0.15 ov Cc;, z;, and (6 r )` 1 have a minirnu,u near 'P' -1 300'K whioh
can be explained by the competition of two proccutosi rt.'cr?nmuing of the cuherr,nt eln,+tlc !icnt-
tering and increasing of the coherent inel++atic one with the temperature increase. mince ur.y-
gen rand beryllirun In Be0 are anisotropic and the incohr-rent ocoLterln.t; in Ile dee1'endir4; on the
spin is small, the measured 6t can tin practically divided only In two tnrmai the olmutlc and
inelastic coherent scattering cross-section, d, t to mainly due to the latter at E - 0.0035 cv.
The change of te It with the temperature In the rong;e 290-150)?K can lead to the decr?eanlnt;
of LZ by about 20%.
The sample density was 2.80-2.65 (;/cm3, I.e. it was lose than the theorrLlcol value
3.04 g/cm3. Therefore the question appeared whether the scattering in small ou(;lee, cnriru'd by
the refraction and the diffraction of the neutron wave.:i at the boundarion between uubntunue
and air in the sample pores, effects on the mom:;ured (o f . ndrlitionnl meanurornontrr on Ile')
sample in the form of the fine-grained duet showed that the oacttoring in small singlet art be
neglected.
2. Bte,dy of thermal neutron diffusion In b+'ry)lium and
beryllia
The diifueior. parameters of Be and BeO were studied by mennu of the pulse r:e thr,d with the
linear accelerator of IA.E. The method, ne in known, is to measure the damplint; cor,fficlont;
01, of the neutron density in the moderator block with time and to analyse con:+erluently the
dependence of of- upon the block geometry parameter B2:
d. - Y_ C V ..P5,-c./34 (1)
where 5cv is the absorption velocity, D and C are the coefl.lciont?s of diffuel+rci and diffu-
sion cooling respectively.
The coefficient J. was measured for 30 beryllium blocks at 11 = 0.00';-0.11 cm 2 and for
27 beryllia blocks at B` = 0.004-0.095 om Z (see Fig-5)- 'Pile average Be density in blocks was
1.79 g/cm3, that of Boo -- 2.'/9 g/cm3. The memnurements of oC. for nnrch great number of the
blocks allowed to obtain the more nccurato diffusion puramotoro (see Table 11) than those In
C2-8J .
The all experimental values of the coefficient D for Be, Including the ones in Treble I],
coincide within the measurement errors; for BeO the difference is considerably higher than
the error. For the coefficient C the data are in agreement for Roo but as for Be there in
- 2 -
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
no such agreement,
The study of the oeuee of the experimental data diupersion for D and 0 ahowed that it may
be essentially connected with the possible differaaao of the crystal etruoture of the materi-
als under investigation. Thus the estimated calculations of D and 0 for Be0 using the above d3E
for the samples with the grain sires 8J, and 29 j;, g-va the values of D within 10% and the va-
lues of 0 within 606. The offeat of the term with ll? in Eq. I is negligible as it is aeon from
the additional analysis of measurements by means of the computer.
In C 91 for the explanation of the dispersion of 0 it was indicated on the neutron trap
effect which must result in dependence ofd. upon the measurement conditions for email blocks
(the neutron source power, the background level etc.). Our attempts to detect this dependence
for the Be blook of 20 cm3 when the source power changes 10 times and the background level does
3 times failed (Fig.6).
The value of C invaeu?ed give some information on the nout.ron slowing-down near the ther-
mal equilibrium indicating the decrease of the mean logarithmic energy loea~lri He and Ba0
approximately 4-5 times.
3. Measurement of moderation length in Be up to 1.46 ev
and 0.3 ev
The slowing-down density q(r,11.) was measured in the rectangular prism BOx9Ox145 on. The
uz35 converter, irradiutod by the neutron beam of the reactor thermal column, served as a neut-
ron source; the indium foils in cadmium filters were used as the neutron detectors at 1.46 ovj
the small plutonium chhnnbur in the filter with mixture of samarium and gadolinium oxides [10]
was the detector at energy 0.3 ev. The mene+.troment rosults are presented in Fig.7 (for In) and
in Fig.8 (for Pu chamber) .
The measurements by means of the Pu chamber were carried out in cash point under the Jul-
lowing conditions:
1) the chamber was surrounded by the filter of 0.125 K/cm Sm and 0.04 F/cm Gd ('GmOd) 1
2) the chamber was surrounded by the filter of 0.125 g/om2 Bm, 0.(Y4 g/ow2 Gd and 0.35g/cm2 Cd
(l'SmGdCd) 1
3) the chamber without filters (N).
The neutron flux of 0.3 ev was obtained no the difference of the first two meacuroment:n
00.3 = NGmGd - NSmGdCd' and the thermal neutron flux 1PT_ = N - (1-T)-1NSGd, whore T is the
transmission of tho resonanco neutrons by the filtt-r Sm+Od. The inaccuracy of the value T, de-
torminod from tho calculation, effects slightly on 91T , since N,. 4- N.
The curvei for and Y"T, (Fig.:r T) at d.'.atancos t j60 cm from the eouroe are parallel.
It implies tit such distances the denelty of the slowing-down neutrons of 0.3 ev In negli-
gibly small in comparison with the density of neutrons of the same energy in the established
Maxwellian spectrum. Using the ratio ` 103/1'T = (6.58-0.1).10-3 at distances t> 60 cm one can
obtain the slowing-gown density c 3 from the total flux 4103.
From Figs,7, 8 it in soon that at distances 't 30 cm from the nourco the slowing-down
i-ith `t -1}Q+2. cm at '3 L -1.46 ev
density is approximated by this expresuion C~,'t.) . - e-'V
and with ft =91.5 ? 2.5 am at t t = 0. 5 ev. At 2> 30 cm the expression (r`~~ 2 Q L~tlo
valid with A = 7.27?0.10 cm at t^2 = 1.'n6 ov and 0.3 ev; this expression presents evident-
ly the: density of the first collisions. Over the whole region of measurements (t&90 cm) the
kernel of moderation is well approximated by t:he expression (Fig.9)
It (w? E,) - ter, o (2)
~., c 4Yi~l
where IT, = 60cm2, 130om2, 4 = 265cm
P1 - 0.066G, P , = 0.301Z, P3 = 0.027 at S .L = 1.46 er
P1 = 0.541, 11;? = 0.422 , P3 = 0.037 at li ` = 0.3 ov.
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
E-
The square of moderation length (1/6 of the mean square displaoemont at moderatirn) pro-
ved to be equal to Ul (1.46 ev) = 92 { 1.5em2, LR7 (0.3 ev) a 104.5-2.0em2. Their difference
L.(1.46-}0.3ev) a 1.5?2.5 on2 exceeds tli- value calculated by 60% with the assumption that
neutron collides with free atoms. Thus, in the range 1.46-0.3z
a v the atom bond in Bet) la+:tloe
significantly effects on the neutron moderation. The value of L (1.46 ev) is in agreement with
the data obtained 111,12] but it is more accurate. There are no data avntlebla for L j (0.3)
in literature. f
The slowing-down of neutrons ti 2 May in Be and 13e0 was studied on the linear accelerator
of the I" by impulse method, measuring the trannmtsoion for indium, cadmium and samarium
filters (In, Cd, Sm have strong resonances at 1.46 ev, 0.178 ev and 0.0976 ev, respectively) by
a small BP3-counter. In addition to BF3-counter the detector of 0.3 ev neutrons wne also uned.
The detectors were placed in moderator block (Bo-603cm5, Be0-70.00.75 am3) along the neutron
beam axis of the accelerator.
The measurement results for the value of the inverse transmission n Yf)vm time, which
lasted after a neutron impulse moment, are shown in Fig. t). In the same figure the detector
counting rate of 0.3 ev neutrons is given too (Pu?-chamber, ohinlded by HmaCd). In these monmu-
remente the channel width of the time analyzer was from 1~ sec to 10'U. sec.
From the curves of Fig.10 the maxima are clearly seen. For convenience, r '(+) is norma-
lized to upit. The time corresponding to theca maxima in obviously a neutron slowing-down time
up to a given energy, t, , plus the time of flight from the point of the last collision rp to
the neutron absorption in detector ~~ + )/V- where v is the neutron velocity, d
is the mean detector dimension (taking into account the detector hole).
In Table III the values is are given. The errors romult from the uncertainty of the posi-
tion of the maximum, the former is connected with the channel analyzer width and with Wu ina-
ccuracy of the calculation of t/ . The comparison of these values with calculated ones, obtai-
ned for the neutron scattering by Erne atoms shows that the crystical binding, in Be w' !,eO Is
of importance for the slowing-down already. In 1.46-0.3 ev range.
Measuring, with Pu-chamber (Fig.14) made possible to derive another important aloeing-clown
characteristic-time-energy distribution of moderating neutrons of 0.3 ev and to comparo It
with the theoretical one [ 131 t
N Neap `" u) N,, (ul,l) f . (c u,1.)J (3)
(A
*hot" t S I
t
; - 1 --- . ~4)
X0(4{1~y ~
( +
flo(u,t ) -is Ibisson probability of appeeronce of 2j neul.rousduring n time t, if the neutron
appearence at any moment Is equally probable and equal to d ( (4- is lethargy, Ua and
tr are initial and final neutron velocities). The full and dotted lines in Fig.11 correspond
to Poijwon distribution (4) with 2/, = 12 for Be and 18 for BeO,respectively. The die., cepenoy of
theory and experiment at large t iu connected the effect of slowly incroaaing faster 1+E
in formula (55).
From the fact that the dispersion of Poisson distribution equals the nvornje value and
v I 18 the eJ owing down time t
using the relations 1.. and ( t ~.n.~ o ' a
up to the en^rrry 0.3 ev c!- be obtained 1'1,)ependlently of its determination from the poniti'm
nt, f tF
of the maximum (ser, Pubic 1 11. ). Far this purpose the dl oparsion A f = L', ;= ,f- F - f +
(2.9)u. see for Be and 4.3,4. sec for MO) which in conneet:ed with energy range of detnctor sen-
sitivity 1101 , in to be substrnctnd from tf,. Au it "'"At we Jet 1'J.3? sec f or BP and
28,1 sec for NO, and nitro ~ Be- ".19 said ~be, 0rO.12..
The neul;ron slowing-down below 0.1 ov was ntuuied by the moasu.rement of the t,rnnamleulon
- 4 -
'bz
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 of boron filters of 0.012 gfom2 and 0,OP5 glcm2 with I/r-detector, the detectors and filters
being placed outside the block (coo (hie to collimator I only those neutrons may be
detected which were direCted to a block Uurt .eu t" A?' ar4 le > 80?. It simplified significant-
ly the calculation of the filter tranemiunion fl(1) vo the neutron tonporature T. It was used
then for deriving the temperature of moterat..% nuutsuno at the moment t from the transmission
values measured with the anaumpti.on of h!uxuel l Lan distribution (see of gn.l3 and 14, whore To
is the equilibrium temperature at t - +e--.' ). In the calculations of 11 ( r) either theoretioal
re mate for G,L , (E) in Be and RoO [ 14 1 were used or the experimental data obtained in [15]
for 6, in Be and our data for 0,t in BeO with the grain sits 8,,L/.. The full and dotted lines
in Fig-15 a,b correspond to those two ways of calculation of rl (r). It is seen from these
figures that below 0.7-0.5 ov T(t) tends to To exponentially, the relaxation time .4 being
practically independent of neither the filter thickness nor the way of calculation (l~.Beet-
den the data in Fig. 13 and 14 the values 7`- 1 for beryllium blocks of 503 oa3 and or be-
ry'llia blocks of 603cm3 were measured. After correcting for finite block sizes, the thermaii-
cation time FA = 185?20LL see for Be and t tha 204?20A sea for Boo, as a result, was obtai-
ned.
The measurements of is and tth performed, give the possibility for deriving the value
end making the following conclusions. The neutron slowing-down to 1.46 ev in Be and Be0 is
due to collisions with free atoms and lasts leas then 10,aeoo. In the range 1.46-0.3 ev a
brystal binding effect reduced to 0.19310.02 in Be and to 0.109?0.012 in Be0; is in this
,range in 1.5-2 more than to to .46 ev. In the range 0.3-0.5 ev_ below which the spectrum
approximating that of Maxwollian in established, ~~ 0.049?0.005 and ~ Sao 0 0.047-0.005 and
to in 6-7 more than to to 1 .46 ov. At E 6 (9.05-0.025 ev) the moderation in very slow, on an
average, up to 185A see ( in Be ahd up to 204)1: see in Boo. The mean logarithmic energy loan
,in this range for the whole neutron spectrum in dependent on the energy and determined by the
thormalitation times
where E p and Ire are thoi equilibrium energy and velocity.
The energy-time distribution of neutrons in the range to ^. E.* 0.05 ev scan be obtained
from formula (4) by substituting the given values ,. The same values were used for calculati-
on of the Blowing-own area below 1.46 ev (see Table Iy).
The value Li; (0.3 ov) for BeO is in a food agreement with the direct osanurement data
(see f 3).
The moderators containing Be give an additional neutron multiplioatio$ in reactors con-
nooted with the difference of coitributione of (n,2n) and (n, .L ) react loan. As the calcula-
tion of the coofficieat of this multiplication poeeaee.d a large uneertainity[171 we measured
E' for Ee in spherical geometry experiment. The experiment consisted in measurement of in--
tegral neutron fission source power N (fission converter of U235 on the reactor bean), being
shielded by a spherical layer of beryllium or graphite of various thickness (graphite wee
necessary for excluding the variation of the sensitivity of the detector ("allwave" BF3-eoua-
ter) with neutron energy). The presence of a hole in a teryllium and graphite spheres for
transmission of reactor neutron beam into converter caused the anisotropy of fission source
and led to determining the source power N as a result of integrating the counting rate, mea-
suring on a horisontal plane at different angles to the beam axis. For reducing this ani-
sotropy and excluding the corrections for a finite distance from the source to the detector,
the spheres investigated were surrounded by a spherical wood layer 14 on thick.
In Fig.16 the measurement results are shown of No And MB9 vs the mean neutron energy loan
when neutrons paused the graphite and beryllium layers and also K$e INBe(AT)fNo(4& I Is
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
given. The value 4E was caloulutod with the offect of neutron "entengling" in elastic and
inelastic eollicaons with Be and C, and it: c %tuinn sumo Wwortainity which is however of no
importance for KBe (see Flg.17), Curve 2 of gives the more probable run of :revs a
layer thickness. At '12-15 g/cm2 Bo KHu ob'..,iinn the muxtmu* value 1.10?0.015. Using this
value, the multiplication ooofficiont of flaWon neutrons was calculated in BeO too. It was
assumed that eeo~~ - --ft _!t n.D where rL and n. are the fission neutron collision density
K e-I it e oe eto
with beryllium 0101; at' 4iutron nlowiug-down to Do and Boo below the throehuld of Be9(n, 2n),
Bo9(n, d-) and 0 o (n,..) reactions. Thin anawnt'tion in approximately true, nn at E > 4 May
6m 2n, is nearly cenatarit, and at F, L 4 Nov the collision donnitiee in Be and Boo are similar.
Asa result, -- - - 1,34 was obtained. Hence KiscO a 1.08-0,023.
Mae 4.6Z
B. CONTI[IBUTION OF FAST E 'FECTB ON Be TO THE MULTIPLICATION
COEFFIUIENT OF BI,,RYI,LIUM ACOEPRLIE8
A.K.Krasin, O.I.Plindov,
The Inntituto of Enorgetirs of Academy of Bcienoos
D3SH
It is of interest to take into account the effect of (n,20 and (n, q(,) reactions on Be
on critical masses and sizes of physioul beryllium assemblies and to separate the contribution
of fast effect i nt,o the multiplication coefficient.
For this purpose '10-group constants for beryllium were obtained. The Be (n,2n) reaction
has considered as inelastic scattering loading to an additional neutron: The crous-sections of
(n,2n) and (n, .) reactions were taken in from [18_22 J and [23,25 ] , respectively.
For verification of constants obtained the neutron age of fission epeetrum and foot-neut-
ron multiplication ooefficion* in infinite homogeneous beryllium medium were calculated. The
evaluated neutron ago `C = '19 cm2 and the value of fast neutron multiplication coefficient
KBe = 1.087 were obtained, which are in a good agreement with the data `26] and with theore-
tical entimuti.one [1-7,281 , respectively.
With these conotantn the multiplication coofficiouts, critical masses and critical di-
menuloris of pity:;ical az oombl.ios, deucribod in [291 wore calculated in multi-group diffusion
approximation, taking into account thp reactions Be (n,2n) and Be (n, d.) ('a"-cane) and with
no account of them ("b"-case). The contribution of the fast effect on Be was defined as a dif-
ference of thu multiplication coefficients, calculated with the account of Be(n,2n) and Be
(n, oL ) rooctiono or without it.
The r'eeults of calculations are shown in Table V. II: is aeon from Table v that the value
of the fast ofrect on 9-10% Be io somewhat less, than too value 12?4% given earlier for the
same assemblies [291 , but is within the experimental errors. Since the fast effect on Be
was calculated in multigroup theory, the value of the prec"nt work seems more preferable.
Approved For Release
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Ejrlll'.Irc"k;,Y. M.m. loxePyH, M,H.CBAk1coB, A.A,heptluric.H. r, l;t; c)rnn 12, 250 (I962);
M.0.JoxopyH9 H.H.CaAiicos, B.A.Tsp8do11 h'', A.A ATOMHan unoprrln ? 9, 258 (1962);
M.O.Ioxepvll. Anulran euopralrt I9:;
A.A.9epHr~luoB. ATOMIIaH aHePrNft L ~ ' ? m1p,~H, M.p.CoAHicoB, B.A.Tapo6alrbKO,
485 1i96s, , '?i '.~.ti'xepyu. ATOMl{Bn aneprRn J,, 505 (I963)1
M.O.Jexepyri. ATOM11811 UHeprw? ;t , ?.::~r (1Sun1, 4i.~~~7a 3pyt~. May40H9e npollecco eoMONneHwn
110ATp0110B B 6epnnvnii H oioic 6c+pn'u r n a;(yn;, ;;.au >:rTo;iow. He ony6nMHoaoHo. OcHODllue pe3ylr>s-
TBTU RoKnuAHBeETOn Ha le HOBOKOH HOWDOpOP1(nR.
2. A.B.AUIT01IOB A Ap. A xuire "OHa114eoltxe HccneNOBOHHn" MocKBa, Han-Bo All CCCP, 1955, cTp.I58
(AOhnoAw COBOTCH.Ae'JIeroIWn He ucx)1yH8po1HOH nowDopenr n n0 IIHpHOWy RCno)1 aoB811Kp BT(,VHOI
blleyrnK. I(eHeae, 1955).
3. F .0ampbel, P. Stolaon. ORNL, 2076 ('1956).
4. T.Komoto, F.Kloveretrom. Trans. Amer. Nuol.Soc. , 1, 94 (1958).
5. G.de Sauesure and H.G.Silvor. Nuol.Sol.Abotr. 13, 1059 (1959)1 Nucl.Gci.and Fng.,,
195 (1959)s Proc. of Symposium Vienna, 17-20 bc9ober, 1960, p.500.
6. 6.D.B.Iyonger, G.S.Manl, R.Rrxmauina and N.Umakanth Proc.Tnd.ACnd.&'i., ULV, B A, 265
6950-
7. L.S.Kothary, P.G.Khubchandary, React. Gci.and Teohnol., 12, 30 (1961).
S. K.W.Seeman, Nucl. Sci.Abotr., 11, 1407 (1962).
9. G. do ,(aussure. Nucl.Sci. ru:d Eng., '12, 43,1 (1962).
I0. H.Ojexopyu, 11.H.COA1tKOe, A.A.4epnuwlon "flpniopu H TOXrrHKO 3IcdnopRMeIIT8" M? 3, 43 (1962).
II. P.Benoist et al. J1oinaA N? 1192, np(')wTOBnOIIHHH Ipmii Het1 no BTOpyA wexAynopoJHyo roHtbe-
peIHIHIO no tinlpHOMy HrR011BOOBrHR1A 8TOMIIOH uicpcnn (F.enroo, I9`~8).
I2. G.A.Jonnon, A.J.Goodjobn and N.F.Winkor. Nucl.Sci.and Nag., 1U, 171 (1962).
13. l.P.JtnAbrnH, F.II.1)Ta:alfO, ATOM11O11 alicp1'HH, 10, 5 (1961).
14. I(.CnnI'BH, JI.Koxapn. Tpynu B':-,Poll M'Jx/yHapoAlloH itch epollr1HCi no uupuuMy H+;nonbseR1HKU
BTOMHCH atleprfH. leijone, 1'58 P. NadpfHHHO Anlcnn/N HHOC'rPMIIIUx y4rnNX, T.2, CTp.b';"'.
ATOMH3)3JT, MOC IrBQ, 1959.
I5. JIJn.XbiOAc, P.F.IIIDOpq. ATIOJC IIOHTpOHIHrx C8401114H. HBAOHH8 BTOpOe. ATOIAK3,r19T, MoricBO, 1950).
16. )t.loa. liCHTpo11r1He HeCJ1O3ONBHHn HO nAepnnX KOTnOX, CTp.I61. a,9A.HHOt,Tp81rH011 nnTepeTypu,
bloc non, 19514.
17. H.kief,, Nucl. ",ci.and Eng., 10, 83 (1961).
I8. J.S.Marion, J.S.Levin and L.Cranborg Phye.Rev., 114,(1959) p.1584.
I9. G.J.Fio'her,lhys. Rev. 108, 99 (1957)-
20. J.R.Ooyahcr, R.L.Ilenkol, R.A.Nobles and J.M.Kistor. Phys.Rov., ~8, 1'1G (1955)-
21. IU.I'.:iydoB, H.C.JlcoeneB, B.M.Mopoa. CdoplinK "IfetTponHen "8HK8" 1'000TOMn8nfT, n. ,I%I.
22. W.P.Ball, M. Mc:rlregov and R.Booth, 31>,ys.Rev., 110, 1392 (1958).
23. P.11..;telaun and E.C.Campboll, Phvs.Rov., 106, 1252 (1957)-
24. M.E.Battat and F.L.Rioe. 89, 80 (1953)-
25* R.Hr,ao and T.W.Bonnor. Nucl.Ftys.,, 122 (1961).
26. IA.III011Hp0. MaTepNo)IU ICOMHCCHH no BTOMHOH anOprnH CUTA. Hnoprrue peanTOpu, T.1,M.,(195)
27. K.C.Xines and .T.I?,Pollard, Journ.of Nucl.Energy, Part A-B, 16, 7 (1962)
28. P.G.A]lno, P.E.Novak and B.Wolfo. Nuci,Sci.and Eng., 2, N? 4, hetaese (1960).
29. A.I(.I(pOC1Hr, 13.I'.J1yy6oBCKHH, M.ll.JOHgOB, p.lO.Pnnanon P.I(.1'ortwnpun, A.P.Hauuni,
Da, P. B.HaBHJIoi, C. H.OHIOTHH, A.II.COH+rclucon. L'py~ AoxAyHapONKoH I(o,uln(rHTOH" nO r!cnnnr:,
lllr?) 3TOM11ON JHeprnn B MA11pi1HX Ilennx.El I(cr10B8 (I958). AorcnrnA h. 2146.
3620- -7-
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Value of and avern'od ovor the
Vaxaellian spectrum at the noutrun temperature Tn = 3900K
Grain sizes. /L,
advanta-
gieous
average
10-15
8
9.97
9.13
40
29
9.44
8.31
10-20
13
9.56
8.53
-
14
9.70
8,76
Table I
Transfer length, 1.-L,L cm
Diffusion tine] a cosec
Absorption
crone section e at
2200 rn/seo
=y.-7
~- ray
1.46 (In) 7.2
0.3 (Pu) 15.7
0.178 (Cd) 20.4
0.0976 (Sm) 27.6
=0.1
9.3 9.5?1
19.2 26?2
26.3 5113
34.8 88=5
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
10.0?0.4
1746
(1.56-'0.01)?105
(4, 12?0.2? )' 105
29.9'-1.0
1.88-0.020
5.75-'0.020
11.8? 0.4
Table III
Slowing-down time (ink sec) of ti 2 May
neutrons to various onerCiee
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
BloWiag-dowu time and area 4 j of neutrons
up to 1.46 ev in Be and Beu
t
aeo
C?
t0 e+ec _
cm2
2 YeV - 1.46 eV
7.511
85.8?16
9.511
92'+-1.5
2 YeV - 0.3 eV
2 Y
V
17.5?1
91.5?2.4
2712
103.4-'1.9
e
- 0.178 eV
40?3
98.9?26
5113
112
1?2
1
2 Y.V - 0.13 eV
56?4? )
103.4?2.7
69?4?)
.
.
+
2 YeV - 0.0976 eV
2 H
t
73?5
+
107.5?2.9
83t5
117,. 7-:'.4
122.5?2.9
o
- (0.07-0.045)eV
135
160
-
(0.07-0.045) - 0.05 eV
185?20
20425
-
?) It In calculated using the obtained values of in
the energy range 0.3 - 0.0976 ev.
Calculation results of Reff' critical sizes,
critical mauues and fast effect on Be
serf
4a 1,0301
4b 0.9286
5a 1.0341
5b 0.9344
6e. 1.0241
6b 0.9270
Rccil
cm
Rexp Bcal
on kg
37.9 40.4 4.8
48.4
34.7 37.5 5.02
44.6
34.3 36.5 6.10
45.3
Oat 1,0396 29.9 31.5 2.78
4b 0.9447 36.5
5a 1,0431 25.8 28.4 2.79
5b 0.9503 32.5
6a 1.0420 23.9 26,2 2.85
6b 0.9508 30.0
36 2 6L
Bexp
kg
RBe
5.46
0.1016
5.86
0.0997
6.66
0.0971
3.31
0.0949
3.36
0.0928
3.42
0.0912
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
6L,6aps
12r-
10.3 2 3 4 5 678910-2 2 .1 4 5 6 789x0'' 2 3 4 5 678910? 2 3 4 5678910'
E, ),?
Fig.1. Total cross section 6., for the BeO samples of various
crystal grain sizes; sample I i): grain sizes = 4-23, , ave-
rage size = 8),- , advantageous sizes = 10-15,,2 12(9 )
sizes = 9-60p., average one = 29,u, , advantageous one = 40g.;
3 (a ) : sizes = 5-37,w- , average one = 13,&- , advantageous
ones = 10-20)u, ; 4 ('X): sizes are large than ones of I and
are less than ones of 3, average size = 14)1' .
e~A ~r.'~i,... ~. d1ti.Y~.t.l'~t, ~ ~- 1-=fx'~-r9-k;.t?-rrr
Fig.2. Total cross section for samples I and 2 produced by
extrusion (a) and pressing (b). The incident neutron beam is
parallel (') and is normal to direction of extrusion and of
preskng (x). The partial cross sections for the reflection
from the plane (110) (see the insert) differ by 20-25%. The
solid line represeutsthe value of 6 calculated with the
assumption of absence of the extinction and the account of
only elastic scattering.
--I T
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
f, so0 atx
E?opab
2 3 4 5691910'
E, 9e
Fig, 4. Total cross section for BeO vs temperature. and
(6 '' )are averaged over the Maxwellian spectrum for neut-
fScon temperature Tn = T + 100?K.
362'- - 11 -
cc.~DK
12r
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
M
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
p c
0 / T ! V S . 7 1 C Mq~
I e",
060 060 (00 (.X rW r!0 fro
!gym ~
7 d 9 C;
Fig.5. Damping coefficient oL vs B2 for Be (a) and BeO(b).
The dots are the measurement results, the solid line
corresponds to parabola (see Eq.(1) and Table II).
Fig.6. Neutron density damping in Be block of 203cm3.
1,2,3 are various powers of a neutron source; 4-with
the large background level. The dotted line represents
the background plus the effect, the solid lines are ob-
tIned after the substraction of the backgrciaud.
362 C..
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Fig.7. Moderation density at E = 1.46 eV for BeO in verioua
coordinates
N ,omx. ea.
11111 A NI
Fig.8. Measurement results with Pa-chamber: (1) flux of 0.3 ev
neutrons o3 (2) contribution of higher resonances;
(3) thermal, neutron flux ep ; (4) moderation density q at
E = 0.3 ov. I - relative IL-it, II-g, cm
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
,(r)
vv, ,
J.0 `
T7F
t
IN, I
A
%
~ r
Fig.9. Comparison of the slowing-
down densities at 0.3 ev(o) and
1.46 ev(e).The solid and dotted
lines correspond the synthetic
nuclei of type (2) at Er=0.3 ev
and 1.,46 ev.
t
o ro m b ro so Mw
09
44
I t 1
4s~
uaj
41
40o
4R'I r' ~ .nce~
i ! o fi7i9ro to !0 wSDiofO.I OW AV XW
Fig.10. Measurement results of the
slowing-doom time. 1.2 are the de-
tector counting rates H(t) of 0.3ev
neutrons in blocks of Be and BeO,
respectively; the inverse transmis-
sion n"+(=) of Cd(3,4),Sa (5,6) and
In(7,8) filters in Be and BeO.
7
6
Fig.11. &zpe=inental energy-time
distributiaa of slowing-down neut-
rons at E =0.3 ev in comparison
with the theoretical one. ~ ? -ex-
periment for Be and BeO. The solid
and dotted lines corres and to Fo a-
son d36tribution : N Be ti 4L
11 11 ) e
N E'Y'e. rt~
Fig. 12.Experimental arrangement
for studying of transmission thro-
ugh boron. 1-neutron collimator
with Cd and ByC; 2,3-filters; 4-
moderator block; 5-collimator sup-
port; 6-collimator holes; 7-BF -
counter; 8-shield of counters.3
Approved For Release 2009/08/26: CIA-RDP88-00904R0001
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
Fig-13t Trensiission of boron filtes for the Be block
60 cm ; I-filter 0.012 g/cm ; 2-filter 0.023 g/cm2;
d ~ ? -various measurement series.
Fig. -14.Tranemission of boron f-l-
ters for BeO block 70XSOX?5 cite
I-transmission; II ,1t sec.
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8
0011 I t
0 : j: r6D 64D SOD X V /BO Q.$D -Ce.
Fig.16.I tegral counting rate N vs -
the average neutron energy lossAE:
1 - Be layer; 2 - graphite layer;
3 - Kg N~(eF-)/IN,~,j E); 4 - gBe
with correction of hole effect in
spheres or the transmission of
neutron beam
? ly'! ND 6`D SX KD
d
Fig.17. 'Effect ofdf calculation
inaccuracy upon EBe ' The calcu-
lation variants: 1 -with utiliza-
tion of cross-sections on 15% as
large in comparison with the
data D5]; 2 - with utilization
of Gros-sections ni 5; 3 - gBe
with the account of neutron ab-
sorption in a wood sphere;
4 - with an assumption that6;Z4_ ,.
1 - Beryllium ~ayer thickness,
g/cm
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110010-8