SOVIET ATOMIC ENERGY VOL. 54, NO. 1

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP10-02196R000300020001-9
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 88
• Document Creation Date: January 4, 2017
• Document Release Date: February 6, 2013
• Sequence Number: 1
• Publication Date: January 1, 1983
• Content Type: REPORT

Ueclassifled and Hpproved For Kelease ZU13/UZ/U6 : GIH-KUF'1U-UZ1y6K0003UUUZUUU1-y
..
- ~..
l ~ E ?
Russian Oiigirial Vol. 54, No. 1 January, 1983 ~ ~ ,
,y .
ATOMHAA 3HEPf NA
:(ATOMNAYA ENERGIYA),
;~ .
_,
TRANSLATED FROM RUSSIAN
July, 1983
~:, '
CONSULTANTS BUREAU, NEW YORK
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
SOVIET
ATOMIC
ENERGY
Soviet Atomic Energy is abstracted or in-
dexed in Chemica/ Abstracts, Chemical
,Titles, Po/Iution Abstracts, Science , Re-
search Abstracts; Parts A' and B, Safety
Science Abstracts Journal, Current Con-
tents; Energy Research Abstracts, and
Engineering Index.
Soviet Atomic Energy is a translation of Atomnaya Energiya, a
-.- publication of the Academy of Sciences of the USSR.
An agreement with the,Copyright Agency of the USSR (VAAP)
makes available both advance copies of the Russian,journal and
original glossy photographs and artwork. This,serves to decrease
the necessary time lag between publication`of the original and
publication of the translation and helps to improve the quality
of the latter. The translation began with the first issue of the
Russian journal.
i
I. N. Golovin ~
V. I.II'ichev,
V. F. Kalinin -
P. L. Kirillov
Yu. I. Koryakin
E. V. Kulov ~
B. N. Laskorln
Editorial Board of Atomnaya ~nergiya:
Editor: O. D. Kazachkovskii
~ ,
Associate Editors: N. A. Vlasov and N. N. Ponomarev-Stepnoi'
Secretary: A. I. Artemov~
- ,
V. V. Matveev
I'. D. Niorokhov
A. A. Naumov.
A. S. Nikiforov
A. S. Shtan'~
B.,A. Sidorenko
- ;M. F. Troyanov
Copyright ?1983, Plsnum~Publishing Corporation. Spvier Atomic Energy partici- ;
pates in the program of Copyright Clearance Center, Inc. The appearance of a
code line at the bottom of the .first page of an article in this journal indicates the
copyright owner's consent that copies of the article may be made for persAnal or
internal use. However, this consent is givc^ on the condition that the copier pay the
stated per-copy fee through the Copyright Clearance Center, Inc. for all copying not
explicitly permitted by Sections 107 or 108 of the'IJ.S. Copyright Law. It does not
extend to other kinds of copying, such as copying for general distribution, for
advertising or promotional purposes, for creating new collective works, or for resale,
nor to,the reprinting of figures, tables, and text excerpts. ~ -
I
Consultants Bureau journals appear about six months after the publication of the
original Russian issue. For bibliographic accuracy, the English issue published by ,
Consultants Bureau carries the~same number and date as the original Russian from
which it was translated. For example, a Russian issue published in December will
appear in a Consultants Bureau English translation about the following June, but the
translation issue will carry the December date. When ordering any vblume or particu?
lar issue of a'Consultants Bureau journal, please specify the date and, where appli-
cable, the volume and issue numbers of the original Russian. Tite material you will
receive will be a translation of that Russian volume or issue.
Subscription (2 volumes per, year) -
Vols. 52 & 53: $440 (domestic); $489 (foreign) Single Issue: S 100
Mailed in the USA by Publicatjons
Expediting, Inc., 200 Meac_ ham Ave-
nue, Elmont, NY 1 1003.
POSTMASTEfi: Send address changes to
Soviet Atomic Energy, Plenum Publish-
ing Corporation, 233 Spring Street, New
York, NY 10013. '
Vols. 54 & 55:$500 (domestic); $555 (foreign)
i
CONSULTANTS BUREAU, NEW YORK AND LONDON
n
0
233 Spring Street " "-
?New York, New York 10013
%,
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
SOVIET ATOMIC ENER ~(
G
A translation of Atomnaya Energiya
July, 1983
Volume 54, Number 1 January, 1983
CONTENTS
Engl./Buss.
ARTICLES
Temporal Evolution of the Neutron Spectrum in a Coherently Scattering
Crystalline Moderator of Small Size - Sh. Kenzhebaev. 1 14
Transfer and Deposition of Radioactive Nuclides in a Convection Flow
of Sodium - A. S. Zhilkin, A. P. Kondrashov, E. S. Kononov,
A. A. Kutuzov, and 0. V. Starkov 5 17
Angular Distribution of a Broad Beam of Fast Electrons Reflected
from a Semiinfinite Medium in the Case of Grazing Incidence
V. S. Remizovich 10 20
Effects of Radial Diffusion on the Possible Erosion or an Open-Trap
Plasma - V. G. Petrov 14 23
Equipment Complex for Monitoring the Neutron Flux of the~Control and
Safety System of Water-Cooled/Water-Moderated Power Reactors of
Nuclear Power Stations - G. F. Borovik, I. E. Burenko, A. M.
Gusarov, V. S. Zhernov, M. S. Kalenskii, I. S. Krasheninnikov,
V. A. Meshkov, Yu. B. Prokhorov, and A. G. Yakushev 19 27
Induced Activity of Certain Concretes Irradiated at a 680-MeV
Proton Accelerator - V. F. Kas'yanov, A. N. Kargin, M. M.
Komochkov, B. V. Man'ko, and B. S. Sychev 31 36
Radiation Conditions in a 16-MeV Electron Microtron Accelerator -
A. G. Below, G. A. Komendantova, Yu. G. Teterev, and
A. P. Cherevatenko 35 38
Electromagnetic P4ass Separator for Radioisotope Separation - R. I.
Lyubtsev, V. I. Orlov, V. S. Belykh, A. G. Evdokimov, V. N.
Voichishin, G. A. Akopov, V. Ya. Mishin, B. I. Rogozev,
M. K. Abdulakhatov, and E. M. Rubtsov 42 43
Determination of the Coefficient of Isotope Separation in Chemical
Exchange by the Method of Multistage Extraction - S. D. Moiseev,
V. A. Samoilov, and Yu. I. Ostroushko 46
Depth Distribution of Hydrogen in Metals by the p-p Scattering Method
V. N. Kadushkin, Z. P. Kiseleva, G. A. Radyuk, B. G. Skorodumov,
I. I. Trinkin, V. A. Shpiner, P. K. Khabibullaev, and V. N.
Serebryakov 50 49
LETTERS TO THE EDITOR
Apparatus for Determining the Direction of Flow of Underground
Water Revealed by a Drilled Well - I. G. Skovorodnikov 56 54
Effect of Boration on the Activity Induced in Concretes by Proton
Bombardment - V. F. Kas~yanov, A. N. Kargin, M. M. Komochkov,
and M. F. Mitin 59 56
Radiation Creep in 09Kh16N15M3B Steel at Stresses Exceeding the
Elastic Limit - A. S. Kruglov, V. N. Bykov, and Yu. M. Pevchikh 62 57
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
GVNTCI~I~S
(continued)
Engl./Buss.
Plastic Scintillators for Recording fast Neutrons - D. V. Viktorov,
L. A. Gorbunov, I. M. Rozman, A. M. Sirenko, and V. M. Shoniya
of Utilization of the Neutron Flux in Netron-
c
i
ffi
64
58
en
y
c
Increased E
f the Californium
i
ng o
Activated Setup with Centralized Position
67
60
Source - B. S. Vakhtin and G. A. Kuznetsov
Additional Radiation Factors in High-Current Electron Accelerators
Komarov,
L
P
da
Z
.
.
,
ava
L. F. Belovodskii, V. D? Egerev, N. I.
N. A. Mishin, A. V. Pilipenko, and M. D. Volodin
Helium Blistering of Nickel with a Temperature Gradient in the Surface
nenko
Mart
V
~0
62
,?
y
.
Layer - M. I. Guseva, S. M. Ivanov, Yu.
72
63
and A. I. Ryazonov.
Taking Account of the Background of?Natural Neutron Radiation in
Determining the Composition of a Mixture of Fissile Nuclides from
Nesterenko,
S.
V
onov
Mi
?
?
:
,?
r
Delayed Neutrons - B. P. Maksyutenko, A. N.
and Yu. F. Balakshev
Using the Variational Method of Calculating Plasma
Equilibrium in a Tokamak for the Consistent
Solution of Problems of the Evolution of
Kolesnikov
K
-
75
65
..
.
V.
Equilibrium and Heat Transfer
and V. D. Khait,
Counter with a Plastic Scintillator for Measuring
High-Energy Neutron Spectra - V. E. Aleinikov,
~~
66
M. M. Komochkov, A. V. Solodilov, and G. N.
Timoshenko
gp
68
The Russian press date (podpisano k pechati) of this issue was 12/30/1982.
Publication therefore did not occur prior to this date, but must be assumed
to have taken place reasonably soon thereafter.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 ARTICI Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
TEMPORAL EVOLUTION OF THE NEUTROIQ SPECTRUM IN A COHERENTLY SCATTERING
CRYSTALLINE MODERATOR OF SMALL SIZE
Sh. Kenzhebaev UDC 621.039.51.12
The damping constant ao of the neutron distribution from a pulse source in limited vol-
umes (assemblies) of uniform material at an asymptotically long time t after a neutron pulse
cannot exceed [1]
7~* = min w(v) =1im w(v),
where w(v) is the total probability of interaction of neutrons with velocity v in the mater-
ial. For semicrystalline coherently scattering moderators this limit is attained in compara-
tively large assemblies, e.g., for beryllium with a geometric parameter Bz = B'C2 = 0.03 cm s.
At the same time, a practically exponential decay of the neutron flux with an effective con-
stant a ~ (2-3)a~~ has been observed experimentally [2] in moderator assemblies of significant-
ly smaller sizes.
A number of papers [3-17] have been devoted to the experimental and theoretical investi-
gation of the evolution of neutrons in beryllium, in which it has been noted, e.g., that not
a true asymptotic neutron distribution was observed [2], but some intermediate relatively sta-
ble distribution cahich has received the designation of "quasiasymptotic." In particular, it
has been shown experimentally [5] that in a beryllium assembly with BZ = 0.075 cm_z the neu-
tron spectrum has still not become asymptotic at t = 649 usec.
The onset of the quasiasymptotic distribution and its subsequent transition to an asymp-
totic distribution has been discussed theoretically in [12, 14, 16, 17]. In particular, it
has been shown [17] that in beryllium assemblies with geometric parameters Bz > 2B*2 a stable
quasiasymptotic distribution is not established in general. This phenomenon has been inves-
tigated experimentally and calculated theoretically in [18]. The calculation was done by
the multigroup method using a realistic model of the scattering kernel. The experimental de-
cay rate of the quasiasymptotic distribution turned out to be appreciably lower than the
theoretical rate. One of the possible reasons for this may be associated with not taking suf-
ficiently accurate account, in the multigroup calculations, of the region of extremely small
velocities (e.g., see [18]), at which the singular part of the asymptotic spectrum is formed.
This question has been investigated in this paper on the basis of the simple model of
the scattering kernel of Corn gold-Durgun [12], who propose an analytic investigation of the
neutron spectrum as v -} 0. This model satisfies the principle of detailed equilibrium and
gives a correct total scattering cross section.
Statement.of the Problem. We shall discuss the temporal evolution of the velocity dis-
tribution of the neutrons in a small assembly (B2 > Biz) of beryllium generated by a pulse
source. After cessation of the pulse, the neutron distribution N(v, t) is described in the
diffusion approximation by the equation
(a -{- A > N (v, t) = 0; - ~ ws (v' --> v) N (v', t) dv';
at
0
AN (v, t) = I w (v) -]-- D (v) Bz] .~' (v, t) w (v) _ u'a (v) -f- wt~ (v) -I- wet (v)
where wa(v) = const, win(e), and wel(v) are the probabilities of absorption, inelastic, and
elastic scattering, respectively; D(v), diffusion coefficient of the neutron (in the gener-
al case wel(v) and D(v) are nonmonotonic functions with a finite number of bounded discontin-
uities); and ws(v' -~ v), scattering kernel - the probability of the fact that a neutron with
Translated from Atomnaya ~nergiya, Vol. 54, No. 1, pp, 14-16 ,January, 1983. Original
article submitted February 8, 1982,
0038-531X/83/5401- 0001$07.50 ? 1983 Plenum Publishing Corporation 1
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
velocity v' falls into a unit velocity interval near v per unit time as a result of collisions
with moderator nuclei.
Following Corngold and Durgun [12], we represent the scattering kernel of Eq. (1) in the
ws (v' -~ v) - bwin (v) win (v') M (v) ~- wee (v) S (v'-v),
where b-1 = I w~? (v) M (v) dv, M (v) = 4/V~t v2 X exp ( v2) ; v = E k .
0
We used the inelastic scattering and transport cross sections from [19] in the calcula-
tions, which describe completely satisfactorily the interaction of neutrons in beryllium both
above and below the Bragg limit.
We solved Eq. (1) by the time-step method [20-22]. To this end it was rewritten with
the expression (2) taken into account in the form
N (v, t -[- 0t) _ {1- 0t [wl? (v) ~-- D (v) BZ]} N (v, t)+.Otbwl~ (v) M (v) ~ wln (v') N (v', t) dv'.
u
We determined the effective damping constant a(t) from the formula
~, (t) _ - dN (t)lN (t) dt,
where N (t) = I N (v, t) dv
0
We did the calculations for Bz = 0.075, 0.076, and 0.101 cm 2. We took the following
as the initial velocity distribution of the neutrons:
the experimental spectrum of Gaertner et al. [5] at t = 176 usec in a beryllium assembly
with BZ = 0.075 cm 2; and
a Maxwellian distribution with temperature 600?K and BZ = 0.076 and 0.101 c~.z (beryl-
lium assemblies with dimensions of 15 x 19 X 20 and 15 x 15 X 15 cm). The steps in t and v
were chosen equal to 4t = 1 usec and Dv = 0.05vT for p 5 v 5 3.5 where vT = 2~,
Comparison of the Theory with Experiment and Discussion of the Results. The results ob-
tained are presented in Figs. 1-3, correspondingly, at relatively short and long times after
a neutron pulse. The measured spectra of Gaertner et al. [5] and the computational results
based on multigroup diffusion theory produced by Ghatak et al. [23] using the first term of
the expansion of the scattering kernel of Plachek and the transport cross section taken from
Goyal's research [24]%~ are presented in Fig. 1,
It is evident from Fig. 1 that on the whole, notwithstanding the rather complicated form
of the spectra at a relatively short time after a pulse, the calculations performed in this
paper using a dimple model of the scattering kernel are in satisfactory agreement both with
experiment and with theoretical calculations performed using a more realistic model. The
calculations satisfactorily describe the position and amplitude~of the maxima and minima of
the neutron spectra, which are located at the positions of the maxima and minima of otr(E),
and in particular the fact that as the time increases there occurs an accumulation in the
neutron spectrum at an energy corresponding to the maximum peak of the elastic scattering
cross section. Such agreement is evidently explained by the use of the actual total inelastic
scattering cross section and the transport cross section.
The accumulation of neutrons in the region of small energies (in the so-called Corngold
"trap") predominates at large times in accordance with the prediction of theory [16] that
in a small moderator volume an energy spectrum of neutrons asymptotic in the time has a sin-
gularity at E = 0 of the S-function type. This is evident from Fig. 2 for B2 0.101 cm-2.
However, one should note that the shape of the neutron distribution in thin energy re-
gion, according to our calculations and the calculations of Japanese physicists [18] (.Fig. 3),
*Since the spectra in [5] were normalized arbitrarily, their normalization was chosen from
the condition that they agree with experiment at E = 500?K for comparison with our results.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
0,9 ;0 v= f rcT
Fig, l
f0''2
f0-f5 [
0
~0 2, 0 V= f rcT
Fig. 2
Fig, 1. Time-dependent neutron spectra. at a short time after a pulse in a berylli-
um assembly with B2 0.075 cm-2 at t = 176 (1), 378 (2), 477 (3), and ~ (4) usec:
p , x) experiment of [5]; -----, ) calculation of [23]; ?) calculation of the
present paper.
Fig. 2. Time-dependent neutron spectra at a long time after a pulse in a beryllium
assembly with B2 = 0.101 cm 2, based on the calculations of the present paper at
t = 1 (1), 2 (2), and 4 (3) msec.
differs appreciably. This may be related both to the inaccuracy- of-the multigroup calculation
in [18] and to the approximate nature of the scattering kernel used in our calculation. Not-
withstanding this discrepancy, the time dependence of the effective damping constant a(t) of
the neutron flux according to our calculations and the calculations of the Japanese physi-
Gists [18] proved to be similar. This is evident from Fig. 4, in which the results of both
calculations are presented as well as the experimental data o~ [18]. It is evident that ac-
cording to the calculations and the experiment with BZ = 0.076 cm_2 there exists: a rather broad
time interval after a pulse during which a(t) ~ const > a'~. In this interval an almost sta-
ble quasasymptotic neutron spectrum in formed. At large times the quasasymptotic distribu-
tion decays, changing into a purely asymptotic one, and ~(_t) slowly decreases, tending to
~* - its own limit.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
z
Fip? 4
Fig. 3. Comparison of the time-dependent neutron spectra at a long time after a
pulse in a beryllium assembly with BZ = 0.101 cm 2 at t = 4 (1), 2 (2), and 1 (3)
msec: -----, -?-?-?-, ) calculation of [18]; ~ x, +) calculation of the
present paper (both calculations are normalized at E = 0.0.49 eV).
Fig. 4. Time dependence of the effective damping constant of the neutron flux in
comparison with experiment (0 ,?) and calculation [18] for beryllium assemblies
with the dimensions 15 X 15 x 15 (1) and 15 x 19 x 20 (2) cm: -~-?-?-) calcu-
lation of [18]; -----) calculation of the present paper.
The results obtained show that taking correct account of only the behavior of the neutron
distribution in the energy region close to zero evidently can scarcely explain the observed
discrepancy between the theoretical and experimental decay rates of the quasiasymptotic neu-
tron spectrum. This discrepancy is most likely explained by the incorrectness of the diffu-
sion approximation to describe the nonsteady transport of neutrons in small assemblies of
beryllium below the Bragg limit.
1. N. Corngold, Nucl. Sci. Eng., 19, 80 (1964).
2. I. F. Zhezherun, At. Energ., 16, No. 3, 224 (1964).
3. E. Silver, in: Proc. BNL Conf on Neutron Thermalization, BNL-719, Vol. 3, 981 (1962).
4. R. Fullwood et al., Nucl. Sci. Eng., 18, 138 (1964).
5. R. Gaertner et al., in: Pulsed Neutron Research, Vol, 1, IAEA, Vienna (1965), p. 483.
6. V. I. Mostovoi et al., ibid., 1, 623 (1965).
7. L. Kothari, Nucl. Sci. Eng., 23, 402 (1965).
8. A. Ghatak et al., J. Nucl. Energy, A/B, 19, 679 (1965).
9. I. Goyal et al., J. Nucl. Energy, A/B, 20, 667 (1966).
10. J. Wood, J. Nucl. Energy, A/B, 20, 649 (1966).
11. R. Lee and P. Daitch, Nucl. Sci Eng., 28, 247 (1967).
12, N. Corngold and K. Durgun, Iducl. Sci. Eng., 29, 354 (1967).
13. J. Wood, J. Nucl. Energy, 22, 525 (1968).
14. M. V. Kazarnovskii et al., in: Proc. Symp, on .Neutron Thermalization and Reactor Spec-
tra, Vol. 2, IAEA, Vienna (1968), p. 331.
15. I. F. Zhezherun, ibid., p. 379.
16. M. V. Kazarnovskii et al., in: Theoretical and Experimental Problems of Nonsteady Neu-
, trop Transport [in Russian], Atomizdat, Moscow (1972), p. 46.
T7. R. Conn and N. Corngold, Nucl. Sci. Eng., 37, 85, 94 (1969).
18. 0. Aizawa et al., Nucl. Sci. Eng., 50, 38 (1973).
19. R. Bhandari, J. Nucl. Energy, 6, 104 (1957).
20. E. Barnard et al., in: Proc. Brookhaven Conf. on Neutron Thermalization, Vol. 3, 805
(1962).
21. M. Ghanian and P. Daitch, Nucl. Sci. Eng., 19, 343 (1964).
22. A. Ghatak and H. Honeck, Nucl. -Sci. Eng., 21, 227 (1965).
23. A. Ghatak et al., in: Proc. Symp. on Neutron Thermalization and Reactor Spectra, Vol. 1,
IAEA, Vienna (1968), p. 223.
24. I. Goyal, J. Nucl. Energy, A/B, 19, 103 (1965).
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
A. S. Zhilkin, A. P. Kondrashov, E. S.'Kononov, UDC 621.039.553.36
A. A. Kutuzov, and 0. V. Starkov
The problem of the mass transfer of radioactive corrosion products in a sodium circuit
has been revealed and finally formulated within the last decade. If the reactor operates
with pressurized fuel elements, then the radiation environment during maintenance operations
is determined by the radioactivity of, the steel corrosion products. The radioactivity of
54Mn is predominant for all operating fast reactors, The 6OCo radioactivity also makes a sig-
nificant contribution to the dose intensity. in the vicinity of the plant of the primary c.ir-.
cuit.
For such fission products as cesium and iodine, various means of purification have been
developed and introduced, but at the present time it is not exactly clear how to localize or
remove the corrosion products from the circuit. It should be mentioned that considerable at-
tention is being paid in the Soviet Union and abroad to this problem, and investigations are
being conducted in loops and reactors in order to study the mass transfer process and methods
of purifiying the primary circuit of the reactor from steel corrosion products.
A procedure anal the results of experiments on the mass transfer of manganese and cobalt
in a convection sodium loop (CL) are described in the present paper. The results are compared
with the results in other papers.
A diagram of the convection loop designed for studying the mass transfer in sodium of
the radioactive corrosion products 'S4Mn and 6OCo is shown in Fig. 1. Two experiments with
5`Mn and fi?Co were conducted (Table 1), Eaeh experiment consisted of two stages: in the first
stage, a radioactive source was installed in the loop circuit; in the second stage, the source
was removed, the loop was operated without the source, and the redistribution o~ these iso-
topes throughout the loop circuit was recorded.
The sources were prepared in the following way. Specified portions of solutions of man-
ganese and cobalt chlorides, containing 54Mn and 6pCo, were applied.to plates of stainless
steel and nickel with a size of 30 .x 10 x 0.5 mm. Diffusion annealing of these plates, coup-
led in pairs by the active sides, led to penetration into them of manganese and cobalt. to a
depth of ~10 um.
In order to study the dynamics of the removal of manganese and cobalt from the source
in sodium, and their deposition in the circuit during operation of the loop, the activity of
the source and the deposits of manganese and cobalt were measured in individual sections of
the circuit, including the SOCT. After shutting down the loop, the distribution of manganese
and cobalt over the length of the circuit, and also their content in samples of sodium, were
determined in detail. A y-emission spectrometer, based on a NaI(T1) scintillator with a size
of 40 x 40 mm, was used for the measurements. The total error of the measured quantities
amounted to 10-15%.
An important characteristic of the deposition and erosion processes in sodium circuits
is the content of oxygen in the sodium. In the experiments being described, it was estimated
starting from the temperature of the coldest section of the circuit, the SOCT, and the solu-
bility of oxygen in the sodium. For the first and second experiments, it amounted to 12.0
and 6.3.10-6, respectively.
The results of the experiments are given with a correction for the radioactive decay of
5`i~In and 6OCo, Nonlinear effects in the time dependences can be caused by both a change of
Translated from Atomnaya ~nergiya, Vol. 54, No. 1, pp. 17-20, January, 1983. Original
article submitted April 16, 1982.
0038-541X/83/5401- 0005$07.50 ? 1983 Plenum Publishing Corporation S
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
TABLE 1. Conditions for Conducting the Ex-
periments
First experi-
ment (54Mn and
60Co in stain-
less steel)
Second experi-
ment (~Mn in
nickel)
Temperature at heater
outlet, ~
700
700
Temperature drop over
length of the circuit, ~
165
160
Protective g2s
Argon
Argon
Duration of expperiment
with source, h
954
1536
Duration of experiment
without source, h
1360
1056
Material of loop pipeline
12Kh18N10T
12Kh18N10T
Sodium flow velocity,
m/sec
0,0300
0,0300
L MB f source activity,
4
3,1100 (60Co)
0,7930 (54Mn)
Volume of sodium (total)
0,0030
0,0022
ms
Volume of sodium in so-
dium oxide cold trap
(BOGY), ms
0,0009
0,0001
Volume of sodium in
compensation tank, ms
0,0010
0,0010
Total length of loop cir-
cuit pipeline, m
Diameter of pipeline. m
3,5000
0,0200
3,5000
0,0200
20 40 60
Time, days
Fig. 1 Fig. 2
Fig. 1. Diagram of the convection loop: 1) source; 2) hermetic valve for i?nSert~
ing the source and the sample-observers into the loop circuit; 3) cotnpensation
tank; 4) detector with lead shielding; 5) loop circuit; 6) overflow tank; 7) sod-
ium oxide cold trap; 8) heater.
Fig. 2. Dependence of variation of 5L`Mn deposition on the time at a point at a
distance of 0.85 m from the source (1) and in the SOCT (2) for the second ex-
periment (with source).
the rate of removal of the nuclides .from the source, and by their redistribution as a result
of erosion of the deposits from the walls of the circuit.
RESULTS OF THE EXPERIMENTS AND THEIR DISCUSSION
The rate of deposition of manganese decreases as a function of time (Fig. 2). The time
derivative of the buildup curve is large in the initial period of the experiment (250-1000 h),
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
i?++.+a~u .-? i~c laLiVC VV11 LC 11L Vl 1'lll 111 1111
ferent Structural Components of the Convec-
tion Sodium Loop (CL) when Sodium is Drained
Off
*In this experiment a grid of stainless
steel was installed in the compensation
tank.
i
0 ?0 4D 60 BD lOD Time .days
N
4,0
0
~ ?,D
v
w
O 7,0
?~~ 1,5
?p C
6 ~ >.0
0,8
0 0,5 1,0 1,5 7,0 7,5 3,0
Distance from source, m
Fig. 3 Fig. 4
Fig. 3, Dependence of the variation of 54Mn deposition on the time, for the condi-
tions of the experiments described in [1] (T = 650?C): 1-5) points at different dis-
tances from the source.
Fig. 4. Distribution of 6OCo deposits throughout the convection sodium (CL) cir-
cuit (first experiment with source (~j and without source (U)).
but tends to a constant value. The results of the measurements can be described by a para-
bolic function. A qualitatively similar time dependence of the buildup of 54Mn in the cir-
cuit was also observed in other papers [l, 2] (Fig. 3).
The dynamics of the buildup of 60Co in the loop circuit were not investigated in the
present paper, but in [2] it was noted that there exists a certain "induction period" of
1000 h, after which the deposition of cobalt on the walls of the loop circuit increases
sharply.
It can be seen from Figs. 4 and 5 that the distribution of 54Mn and 6OCo throughout the
loop circuit is nonuniform. An increased deposition of manganese is observed in the vicinity
of installation of the device for introducing the source and samples into the loop circuit,
where the temperature of the wall is lower than in adjacent sections of the circuit. The
deposits of cobalt on the walls of the circuit gradually decrease on the whole in the direc-
tion of the sodium stream. When the sodium is drained from the loop, 27% of the manganese
is removed. The residual quantity of manganese remains in the deposits on the walls of the
circuit, the compensation tank, and other components (Table 2).
Knowing the surface area of the circuit and the volume of sodium, the distribution coef-
ficient K can be estimated, which occurs as a constant parameter in the radioactive mass
transfer equation for a uniform circuit [3]. For the quasisteady case, this coefficient can
be defined as the ratio of the concentration of the isotope being studied on the surface of
the loop, to its concentration in the sodium. For the material of the pipeline of the CL
(stainless steel), the distribution coefficients of manganese and cobalt amounted to: in the
first experiment with the source, 1.6 for 54Mn and 0.9 for 6OCo, and without the source, 3.2
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
6 for
0 0,5 1,0 1,5 2,0 2,5 3,0 3,5
Distance from source, m
Fig. 5. Distribution of 54Mn deposits through-
out the convection sodium loop (CL) circuit
(second experiment with source (~) and without
source (p)).
for 54Mn and 1.2 for 6OCo; in the second experiment with the source, 1.7 for 54Mn, and with- k
out the source 5.0 for 54t~1n. It can be seen that the value of K for 54Mn is almost independ-
ent of the oxygen concentration in the sodium during operation of the loop without the source.
This, probably, is explained both by the method of preparing the source and by the conditions
of the experiment in the CL. The. increase of the distribution coefficient in the case of
operation of the loop without the source, by comparison with the experiment conducted with
the source, is related, apparently, with the fact that with the removal of the sodium from the
loop circuit into the overflow tank, a considerable part of the 54Mn is. deposited in the lat-
ter, and the sodium, filling the loop circuit in the experiment without the source, is de-
pleted in this isotope. The increase of K in the second experiment without the source can be
explained by the fact that in this experiment the concentration of 54Mn is less by a factor
of 1.5 than in the first experiment, because of the more intense deposition of 54Mn on the
walls of the overflow tank. The 54P4n distribution coefficient on samples of stainless steel
was measured in [4]. Although the conditions of the experiment were different (the measure-
ments were conducted on the KNK reactor at 200-400?C), the results nevertheless coincide with
those obtained in the present paper.
An analysis of the results obtained in the present paper and in other similar papers
[1, 2, 5, 6] indicates the tendency o? the rate of deposition of sodium to increase in the
loop conditions in regions with a negative temperature gradient. Moreover, the manganese is
deposited also immediately behind the source. The cobalt is deposited predominantly immed-
iately behind the source, and then its concentration on the wall decreases, having tended to~~
a constant value. It was noted in [2] that the distribution of the cobalt deposits along the
length of the loop can be described by an exponential relation. It can be seen from Fig. 4
that the results of the experiment in the conditions of the CL also can be described approx-
imately by an exponent, with the exception of the section of the circuit in the vicinity of
the location of the source, where the distribution of the deposits is almost uniform over the
length of the section.
Consideration of the measurement results in the experiments without the source (Fig. 5)
show that the content of manganese on the wall of the circuit is reduced in the hot section
of the loop but is increased slightly in the cold section. However, variations in the values
of the deposits during operation without the source are small for the CL and, as a rule, do
not exceed 10%. In the compensation tank and the SOCT, increases of the manganese content
were observed by factors of 2.3 and 1'. 1, respectively.
Thus, the results of the experiments carried out showed the following:
1. Deposits of 54Mn on the walls of the sodium loop depend on the duration of the ex-
periment. In the initial period of the experiment (up to 250-1000 h), the rate of deposition
is large, but then it is reduced to a constant value. Obviously, the redistribution of the ~
deposits during-the experiment depends slightly on the time. dependence of the rate of removal
of radioactive nuclides from the source. If this is so, then the behavior of manganese can
be characterized as a diffusion process. In this case, in the initial period of the experi-,;
ment, the rate of removal of this isotope is large, in consequence of its large concentration
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
gradient at the steel-sodium interface; then with reduction of the concentration gradient,
the rate of removal is slowed down. As the experiment shows, the rate of removal of 5`'Mn from
the source with time generally can be stopped, which, obviously, is related with the formation
of a depleted layer in which the duration of diffusion of this nuclide is greater than the
half-life.
2. Manganese has a tendency to increase the rate of deposition in regions with a nega-
tive temperature gradient. At the same time, it is deposited also at the site of location
of the source. Cobalt is deposited predominantly immediately beyond the source. lts concen-
tration on the wall later decreases.
The experiments without the source, conducted after the experiments with the source,
showed that the manganese content on the wall of the circuit is somewhat less in the hot part
of the loop and is increased in the cold part. However, these variations are small and, as
a rule, do not exceed 10%. At the same time, increases of the manganese content by factors
of 2.3 and 1.1, respectively, were observed in the compensation tank and in the SOCT.
3. The distribution coefficient, determined as the ratio of the concentration of the nu-
elide on the surface of the loop circuit to its concentration in the sodium, during operation
of the loop does not vary with increase of the oxygen concentration in the sodium liy a factor
o f two .
1. A. Thorley et al., in: Proceedings of the Seeond International Conference on Liquid Metal
Technology in Energy Production, Richland (1980), p. 13-1.
2. J. Newson et al., in: Proceedings of the Second International Conference on Liquid Metal
Technology in Energy Production, Richland (1980), p. 17x11.
3. A. S. Zhilkin and I. A. Kuznetsov, in: Radiation Safety and Protection of Nuclear Power
Stations [in Russian], No. 5, Atomizdat, Moscow (1981), p. 78.
4. M. Stamm et al., in: Proceedings of the Second International Conference on Liquid Metal
Technology in Energy Production, Richland (1.980), p. 17-58.
5. W, Brehm et al., ''Techniques for studying corrosion and deposition of radioactive mater-
ials in sodium loops," in: Data of an International Conference on the Behavior of Fission
and Corrosion Products in the Primary Circuit of Fast Reactors [Russian translation],
Dimitrovgrad (1975), p. 172.
6? N. Sekiguchi et al., "Behavior of corrosion products from irradiated stainless steel
in flowing sodium,'' in: Data of an International Conference on the Behavior of Fission
and Corrosion Products in the Primary Circuit of Fast Reactors [Russian translation],
Dimitrovgrad (_1975), p. 82.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP1O-021968000300020001-9
ANGULAR DISTRIBUTION OF A BROAD BEAM OF FAST ELECTRONS REFLECTED FROM
A SEMIINFINITE MEDIUM IN THE CASE OF GRAZING INCIDENCE
When thick layers of a material are irradiated with a flux of fast electrons, a signfi-
cant fraction of the electrons is reflected at the medium. The relative number of reflected
electrons increases substantially with decreasing angle ~o between the velocity vector of the
incident electrons and the target surface.
But the existing theories of`electron albedo [1-3] describe grazing incidence only made-
quately when ~o ? 1. The reason is that the main assumption on which the above wmrk.ha.s been
based implies that the main contribution to reflection is provided by single scattering at
relatively large angles ~ z 1. Multiple scattering of electrons before their entry into the
medium and before large-angle scattering events, as well as after such scattering and before
the backward exit of an electron from the material, were completely disregarded in [1] and
were considered second--order processes in [2, 3], with the processes "smearing out''' the over-
all pattern of angular scattering and energy scattering of the reflected electrons. It is
quite clear that in "grazing'' incidence of a particle beam on the surface of a material, the
multiple scattering is not a second-order process at ~o ? 1, but is the main process, because
in this case the greater portion of the electrons can be reflected from the material without a
single large-angle scattering event. The greater part of the electrons can be simply reflec-
ted by multiple scattering under relatively small angles. Disregarding this fact implies a
substantial disagreement between the theory [2, 3] and the experimental results [4]. We con-
sider below the problem of calculating-the angular distribution of reflected electrons in the
irradiation of a semiinfinite inhomogeneous target medium by a broad flux of fast monoenerge-
tic electrons propagating in a single direction. We restrict our considerations to the case
of purely elastic scattering. This situation is encountered in reality when a large fraction
o? the electrons has time to leave the medium before losing some noticeable part of the ini-
tial energy, i.e., when w cv 1 applies for the total reflection coefficient.
Assume that a broad monoenergetic beam of fast electrons is incident under a small angle
~o onto the surface of a plane layer of an inhomogeneous material. Assume the z axis to be
parallel to the normal to the surface and the XOZ plane to coincide with the velocity vector
of the incident particles. The direction of particle motion is defined by the angles ~ and
cp; ~ denotes the angle between the velocity vector and the XOY plane and ~P denotes the azi-
muthal angle, i.e., the angle between the velocity vector and XOZ plane. Values 5 > 0 corres-
pond to particles moving into the material, whereas values ~ < 0 correspond to particles .mov-
ing in the opposite direction. In our case of grazing incidence, the considerations can be
restricted to the small-angle approximation of [5, 6] when we assume that ~~ and ~p~ ? 1,
In this approximation the projections of the velocity unit vector ~ are. related to the angles
~ and ~ by the formulas ~ = cos ~ cos. ~P .:, 1; ~ = cos ~ sin.~p ~ ~P ; ~ = sin ~ ~ ~; the angle
of single scattering fromxthe state ~(_~;cp) into the state n'(~'; ~p')zis given by the formula
~'' = arc cos (~' , ~) = ~/(~ - ~) + (cp' - cp) 2. In the case of fast electrons, the probability
of single scattering is given by the following formula of [7]:
( ) 1 d6 _ Jeff a ^, 1.2.10_SZ2~3 (1)
I S~ ~ 52 = a dS2 ~' '~ ('leff+(~'-~)2+(~P'-~P)21z eff^' Eo (1-I-Ea) '
where'~'eff denotes the "effective" angle of single scattering, related to the Moliere shield-
ing parameter by the formula ~z = 4p; Z denotes the atomic number of the material of the
e~f
scatterer; and Eo denotes the e ectron energy (MeV).
In view of the above considerations, we obtain the following form of the transfer equa-
tion of the electron flux density N(T, ~, ~P) at the depth ti = 6 i no (z') dz' (where T is the
0
Translated from Atomnaya nergiya, Vol. 54, No. 1, pp. 20-23, January, 1983. Original
article submitted February 5, 1982.
10 0038-531X/83/5401- 0010$07.50 ? 1983 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/06 :CIA-RDP1O-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
ial at the depth z):
8N t4Pff (~ d~, (' d~. N (ti; ~~; W')Z ~' (i; ~; ~P)Z .
ai n _~ _,~ ( eff-I-(~~-~) -I-(T~-T )1
Additional conditions to be satisfied by transfer equation (.2) are
AT(i=0; ~>0; ~)=Nos(-~o)s(~P); N(~; ~; ~P)=0 if ti-->oo; ~Z~~Z->oo, C3)
where No denotes the particle flux density in the incident beam. The second of conditions
(3) is the requirement that the solution must remain finite in the depth and over the angles.
The angular distribution of the particles reflected at the material (for such particles
the angle is ~ < 0) is described with the reflection function S (~~I, ~P; ~o), which is re-
lated to the particle flux density at the boundary of the material through the formula
N (r = 0; ~ < 0, ~p) = ATo s (I ~ I , w; ~o)
ICI
There NoS(I~I, cp; ~o) determines the number of particles leaving a unit surface area of the
medium per unit time in the angular interval ICI-ICI + d~~cp - cp + d~Q. Since the number
of particles incident per unit time on a unit surface area of the medium is ~oNo, the differ-
ential reflection coefficient dew is associated with the reflection function by the simple
formula dew = ~o1SdI~Id~P. The reflection function S fully describes the angular spectrum of
the reflected electrons and must be determined during the solution of the problem.
Since the transfer equation is linear, in the case of a semiinfinite homogeneous scatter-.
ing medium the flux of electrons moving toward the boundary, i.e., for ~ < 0, is at any depth
a linear functional of the flux of particles moving into the bulk of the medium, i.e., for
particles with ~ > 0. The kernel of the functional must be independent of depth:
I~IN (~; - ICI; ~) = J d~' ` d~'S (ICI; w' - w; ~') N (~; ~'; ~v')?
-~ o
We took into consideration that after intrpducng the optical depth T in place of z, Eq. (2)
does not differ from the corresponding equation for a homogeneous medium. By assuming T = 0
in Eq. (.5) and recalling the definition given in Eq. (4), we observe that the kernel of the
?unctonal (.5) coincides precisely with the reflection function if ~' _ ~o and ~' = 0. There-
fore, the kernel is denoted by the same letter S.
Important in our subsequent considerations will be the fact that the general solution of
transfer equation (2) can be easily found in the case under consideration if the Fourier
transform over the angular variables ~ and 0). (8)
l eff o 0 o eff o
0
Let us note as an important point that although the fundamental relationship (5) (which is
also the basis of the "invariant inclusion method") was used in deriving Eq. (.8) for the re^
flection function S, the equation obtained is linear, whereas the analogous equation for the
reflection function (equation obtained with the invariant inclusion method) is significantly
nonlinear [1, 7].
When we proceed to the solution of Eq. (8), we note that in the case of electrons with
an energy of tens of kilovolts and more, the effective angles of single scattering are ex~
tremely small. For example, we obtain from Eq, (1) for 100 keV electrons that $eff 10_z
for Z = 1 and' eff '" 4'10_2 for Z = 64. When the particle energy increases, the $ values
decrease even further. Therefore, when the grazing angle ~o is such that the inequality
1? ~o? 'eff
is satisfied, we obtain in the angular range I~Icv ~o of greatest interest that the effective
'~ ff values are weff ? 1 and that the function gKl(q) on both sides of Eq. (8) can be expand-
eed in a series by assuming gKl(q) r 1 + q (ln q + 2C - 1), where C = 0.577 denotes the Euler
constant. We then obtain in the region weff S l;
~ 3 9 - OG
d4 [1-gKi(4)]^ 3 In ~ w$) ti 3 ~w~ ~ (10)
n
where a denotes an approximation parameter varying in this problem within the interval 0
a < 1. The particular a values can be determined either by comparing the results of the
calculation with experimental results or theoretically by employing some additional concepts
which, e.g., can be obtained from the condition that the functions w3 In (1.8/w2) and w3-a
coincide on the average within the interval 0 < w < 1. We obtain a 4/'7 in this determina-
tion. By substituting Eq%((10) into Eq. (8) and introducing the new variables w' = c~aa-3;
e = 1/~~; ~' - (1/'~eff)~`1 s-a , we obtain the following equation for the function S = ~'S
3 3 ul - (' de S(I~I~ ~) ~ dwCOS (w ~` - 3 w3-a) . (.11)
dwcos~w~l~l-[--w / o
0
Even when the integrals of Eq, (11) are not taken over .w , it turns out that the equation is
exactly solved, with the solution being expressed through elementary functions. Indeed, the
Melin transform over the variable a can be used for the solution of the equation [8], By
multiplying both sides of Eq. (11) with as_1 da and by integrating over a within the limits
0 < a < ~ with the aid of the convolution theorem, we obtain an expression for the Melin im-
age of the reflection function S(~I; s). Then, using the conversion equation (8), we obtain
for the reflection function
S(I~I; ~~)_ ~ s((in(ns11) 11 1l
2cos(~~)-~l Cal /g+~ ICI /~J' (12)
3- a
Q=2 4-a .
When a varies from zero to unity, the parameter S varies within the limits 4/3 < (3 < 3/2,
i.e., by a total of 11%. Thus, the form of the angular spectrum of the elastically reflected
electrons does not very strongly depend upon a, This fact a posteriori justifies to some ex-
tent the use of the approximation (.10).
Let us note some general properties of our solution represented by Eq,, (i2):
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
,u r-- ~ ~
0, 9
O.B
~ ,.
~ 0, 6
0, S
z 0, ~
0,3
0,2
0, 7
Fig, 1. Angular distribution of reflected electrons:
calculations made with Eq. (14) f'or the a values 0 (curve 1),
4/7 (curve 2), and 1 (curve 3); ------) result of experiment [4]
on 25 MeV electron scattering at lead with the grazing angle
~o = 20?; -?-?-?-) the same, grazing angle ~o = 40?.
1. For any a the reflection function of Eq. (12) is normalized by the condition
r~
s(I~i; ~o a) d l~l=~o;i.e.~ w= ~o J Shc~l =1. C13)
o u
This condition expresses the obvious fact that the total reflection coefficient w is equal
to unity in the case of purely elastic scattering.
2. Regardless of a, the angular distribution of the reflected electrons has a sharp max-
mum at ICI = ~o, which corresponds to the "mirror" law of reflection. The reflection func-
tion maximum is Smax - -S/2 ctg ~rQ/2, so that the formula for the reflection function normalT
ized to unity at its maximum is
4 sine rzs
S Usl; ~o) _ 2
Sn
=
orm
3 (~o; ro)
2 `11 ~,0 11 ICI JJ
3. When we assume in Eq. (12) that a = 0 (j3 = 3/2), we obtain
S(I~~; So; a=0)=Ln Cll~lla/L+( X0)3/~l-1
~o / l ICI J
Equation (15) coincides exactly with the formula which Frsov obtained in [5] using ttie dif-
fusion approximation over the angles for calculating the reflection function.
4. We obtain from Eq. (12) for a = 1 (.(3 = 4/3):
s(I~I; ~o; a=1)=~~3 C1+~1~1)4/3+/~0)4/3~-1. (16)
V L ~o 11\\ISI
This formula fully agrees with another formula obtained in a later work by the same author
[6), which describes angular spectra of elastically scattered particles when the interaction
of the particles with. the atoms of the material is approximated by a potential inversely pro-
portional to the square of the distance (U (r) r-z).
5. The angular spectrum of the reflected particles is independent of the fprm in which
the density of the medium changes as a function of depth in the medium, i.e., it is independ-
ent of no(z).
6. Formulas ('12) and (14), which we obtained for the reflection function, are universal
functions of the angular ratio ICI/~o and are independent of the energy Eo of the incident
electrons for any a and are also independent of the atomic number Z of the material of the
scatterer.
All these features of the angular distribution of the reflected electrons are in good
agreement with the experimental results [4; 9] for grazing incidence on a thick target. Figure
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
1 shows rerle~~luii iuu~~luua, wiil~u ate Leuu~eu ~u uui~y a~ ~nelr maximum ana wni.cn were cal-?
culated with Eq. (14) for various a values of the interval 0 ~ a ~ 1. Figure 1 also shows
the results of an experiment of [4] on the angular distribution of reflected electrons over
the polar angle ICI; the results were obtained with a lead target at grazing angles ~o = 2Q?
and ~o = 40?. Though the measurements were made on the plane of incidence of the primary
beam, i.e., for ~ = 0, the spectrum differs only slightly from the spectrum of the reflected
electrons averaged over the azimuthal angle, as noted in [9, lOj.
In conclusion, the author thanks 0. B. Firsov and M. I. Ryazanov for their interest in
the present work and for useful remarks. The author thanks Sh., A. Shekhmamet'ev for his help
in the calculations and the drafting of the present paper.
1. R. Dashen, Phys. Rev., 134, No. 4A, 1025 (1964).
2. N. P. Kalashnikov and V. A. P4ashinn, Zh. Eksp. Teor. Fz., 59, No. 6, 2025 (_1970).
3. N. P. Kalashnikov .and V. A. Mashnin, Zh. Tekh. Fiz., 43, 2229 (1973).
4: V. P. Kovalev, V. V. Gordeev, and V. I. Isaey, At. Energ., 39, No. 3, 215 (1975)..
5. 0. B. Firsov, Dokl. Akad. Nauk SSSR, 169, No. 6, 1311 (_1966).
6. 0. B. Firsov, Zh. Tekh. Fiz., 40, 83 (.1970).
7. N. P. Kalashnikov, V. S. Remizopich, and M. I. Ryazanov, Collisions of Fast Charged
Particles in Solids [in Russian], A.tomizdat, Moscow (_1980).
8. G. Bateman and A. Erdely, Tables of Integral Transforms, McGraw-Hill (1966).
9. I. M. Bronsh.tein, V. M. Stozharev, and V. P. Pronin, Fiz. Tverd. Te1a, 13, No. 11., 3359
(1971). ~ T
10. H. Kanter, Ann. Ph.ys., 20, 144 (.1957).
In the design of a reactor having an adiabatic trap with double plugs, it is necessary
to overcome the erosion of the lateral surface of the plasma in the end traps by the surround-
ing neutral gas, this surface being parallel to the magnetic field. Preliminary estimates
show that to prevent this phenomenon the gas-atom concentration should not exceed 108-109 cm 3
[1, 2], which is technically difficult to achieve. In these estimates we have neglected the
transverse diffusion and the thermal conductivity of the plasma, and the condition for stop-
ping the erosion amounted to the number of injected fast neutral particles ionized in the sur-
face layer of the plasma exceeding the loss of ions arising from charge transfer at cold neu-
tral particles in the surrounding gas, while the input of heat with the ionized fast neutral
particles should exceed the energy loss due to ions, electrons, and charge-transfer neutral
particles. The position is serious because the cold neutral particles undergo charge trans-
fer and are ionized in the thin peripheral layer, whereas the source of hot ions is fairly
uniformly distributed throughout the plasma volume. Under these conditions, the plasma may
be disrupted layer by layer beginning with the periphery.
Plasma erosion can evidently be offset by increasing the transverse diffusion coefficient
and thermal conductivity in a sufficiently narrow peripheral region of the plasma in order
that ion and heat balances should occur throughout the region of elevated transport coeffi-
cients, not merely in the region penetrated by the cold neutral particles. Here we consider
the diffusion coefficient required for this and the thickness of the region with high trans-
port coefficients provided that the density of the neutral particles is given at the outer
boundary of the region, where it is assumed that x = 0.
When the thickness of the radial-diffusion region is small, the plasma in it can be des-
cribed by
Translated from Atomnaya Energiya, Vol. 54, No, 1, pp. 23-26, January, 1983. Original
article submitted January 12, 1982.
14 0038-531X/83/5401- 0014$07.50 ? 1983 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
z
-Dax2 =Aexp (-a ~ n"dx) n- i ,
u (1)
where n is the plasma. den.s;_ty; D; elevated diffusion coefficient, which is taken as constant;
x
T, ion lifetime in the trap; Aexp(-v~n dx)n ~ rate of production of hot ions in the layer on
ionization of the injected beam of fast netural particles (the exponential factor incorporates
the reduction in the beam density on passing through the layer); and o = Ko/vo (Ko is the
overall ionization constant for the fast neutral particles caused by ions and electrons, while
vo is the speed of the neutral particles in the beam). The layer is assumed to be optically
thin for the fast neutral particles, and the parameters can be found by assuming that the
S.
value of exp (-6~ n dx) is one. The ion lifetime in the trap before entry to the loss cone is
0
determined by the collisional frequency and is inpersely proportional to the plasma density:
T = nc/An (nc is the plasma density at which the rates of ion generation and loss at the end
become equal). It is assumed that the ion temperature varies little in the layer.
The radial flow of plasma from the central regions is small because of the smallness of
the diffusion coefficient; at the inner boundary of the layer at x = xc, where the diffusion
coefficient increases sharply, we can put
We consider the conditions at the outer boundary of a layer. We first note that the
cold ions produced by ionization and charge transfer from the cold neutral particles are not
retained in"the trap, but are repelled by the ambipolar electric field and automatically enter
the loss cone. They do not have time to interact with the hot ions and their concentration
is small.
We assume that there is a sharp transition from the plasma to the neutral gas having
the minimum possible thickness, which is approximately equal to the cyclotron radius p for the
ions. This requires that the mutual penetration depth for the neutral particles (.before ion-
ization or charge transfer) and the ions due to displacement of the leading center (before
charge transfer) is much less than p. The criteria for this are
Zn 2no~kir+~~ ~~Pe
li=~n~c 0; we use Galerkn's approximate method [3].
We choose the perturbation k in a form satisfying both boundary conditions:
_ / z
k C \ D(1-}-~pxc/D) -{-y-I-Ry2- 1 328 y3 ~ (17)
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
where R and C are arbitrary constants.
The explicit nl(y) dependence is represented as an interpolation polynomial of fifth de-
r~=5
n1 (y) = LJ gyre
r=0
in which the coefficients c are chosen in such a way as to satisfy the exact values o? nl(y)
at six points. The expression for k is substituted into (16), which is multiplied by k and
integrated from 0 to 1. We get a polynomial in powers of a whose coefficients are dependent
on R and RZ. The perturbation can only be a real function, so a, R, and C must be real. We
put ~ = 0 to find the value of R such that the term independent of a is
z ~
r 2pAn1 ( ni) dk r nl l D d k 1 ' r
1-- (0)~-A 1-2- K -I- Z 2 kdy-Aox~ 1 n1 k dy+
L xc ni (~1) 1 no / dy \ nc l xc dy J J~ L
+ xc,:l (u) ay (0) J n1 dy ~- k, J n1 dy] k dy = 0.
The absence of real values of R that satisfy this condition means that the eigenyalue a 0
does not exist and that there is no transition from stability as R varies, i.e., the solution
is stable. Calculations show that. the plasma layer is stable for D S 5 cm2~sec~1, when xc
7.6 cm.
This result, D ~ 5 cm2?sec_1, is about 40 times the classical value. The penetration
depth Z, of the plasma into the gas and ,the same Z for the neutral particles into the plasma
are, respectively, 0.11 and 0.019 cm, and are muchnless than p, which confirms that the above
xc
boundary conditions are applicable. The value exp (-v)~ n dx = 0,945 can be replaced approx-
0
imately by one in deriving the steady-state solution. According to (8) and (9); to retain
n. as T increases, the diffusion coefficient D should increase approximately as I'22 and xc
s~iould increase as r:
There is additional energy dissipation in the surface layer by comparison with the cen-
tral regions, because electron-ion pairs are produced by ionization of cold neutral particles
and these are lost to the ends and take away energy (eu + Te), where a is the potential bar-
rier to the electrons at the end. The following number of cold neutral particles is ionized
per cmz.of the side surface of the plasma:
nrVp kq I, ki
4 ki -I--kt ~ kt
If we assume eu = 8Te for the purpose of estimation, then an additional power input of 2.60
W/cm2 to the surface electrons is required to maintain the heat balance.
Therefore, by increasing the diffusion coefficient in a narrow peripheral region (e.g.,
by producing axial asymmetry or by disrupting the magnetic surfaces) and by supplying add-
tional power to the electrons one can stabilize the side surface of the plasma in the end trap
at a surrounding neutral-gas density of 1011 cm s Such a vacuum is technically feasible.
1. N. Vasl'ev et al., "Tandem mirror hybrid reactor (TROL.project),'' Paper at the IAEA
Working Conference on Particle and Energy Retention in Open Magnetic Traps, Novosibirsk.
(1981).
2. T. Fowler, Plasma Phys., 17, No. 7/8, 583 (1975).
3. L. V. Kantorovich. and V. I. Krylov, Approximate Methods- of Higher- Analysis [~n Russan_],
Fzmatgiz, Moscow (_1962), p. 312.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
EQUIPMENT COMPLEX FOR P10NITORING THE NEUTRON FLUX OF THE CONTROL AND
SAFETY SYSTEM OF WATER-COOLED/WATER-MODERATED POWER REACTORS OF NUCLEAR
POWER STATIONS
G. F. Borovik, I. E. Burenko, A. M. Gusarov,
V. S. Zhernov, M. S. Kalenskii, I. S. Krasheninnikov,
V. A. Meshkov, Yu. B. Prokhorov, and A. G. Yakushev
UDC 621.039.564
Many years of experience in operating nuclear reactors have shown that the monitoring,
protection, and control organization with respect to information about the thermal neutral
flux density (neutron flux) differs in sufficient reliability, rapid response, and accuracy
[1]. Soviet and international rules and recommendations have been developed for the construc-
tion of neutron flux monitoring equipment, defining the requirements and conditions for en-
suring the nuclear safety of nuclear power stations [2, 3]. Taking account o~ modern stand-
ards and recommendations at the start of the 19705, the commercial issue of the third gen-
eraton of neutron flux monitoring equipment (NFME) was developed and brought into use in the
Soviet Union [4]. By means of this equipment, the safety and efficient operation of VVER-44U
and VVER-1000 reactors are ensured in all their operating regimes, including fuel charging
(recharging). The NFME complex (.Fig. 1) shapes the signals which exceed the thresholds (set-
tings) specified by the operator for the power and excursion period of the reactor, distri-
butes signals to the control and safety system, anal processes, records, and presents opera-
tive information at the control desk and panel of the nuclear power station and at the data
computer.
Recently, a work cycle has been completed on the further improvement of the NFME for the
purpose of increasing the reliability, lifetime, and economy, improvement of the operating
characteristics, etc.
The NFME contains a number of subsystems which are functionally autonomous in operation
[4]; fuel charging (recharging) monitoring (FCM), neutron flux monitoring f.or the reactor
control and safety equipment, and neutron flux monitoring from the reserve control panel
(RCP).
At the basis of construction of the complex is the so-called filar structure of the data
measurement channels for shaping the scram system signals, the .warning signaling, reactor
control signals, etc.
The fuel recharging monitoring subsystem contains six detector units (DU), located for
recharging in the reactor core shield. The detector unit signals- are processed by two inde-
pendent and identical subsystems, each of which contains three data measurement channels. The
data are extracted at the unified control panel (UCP) in the central reactor hal]. and at the
recharging machine control desk.
The detector units of the neutron flux monitoring subsystem for the reactor control and
safety system are located in the ionization chamber channels (I C) of the biological shield
of the reactor pit space. Two independent subsystems are provided, which process signals
from the detector units located in pairs in adjacent ionization chamber channels. The as-
semblage of data takes place at three points of the reactor cross section, and ensures satis-
factory representativeness of the data about the neutron flux for subsequent shaping of warn-
ing signals by "2 out of 3" logic [2]. The installed structural superfluity in the form of
two subsystems, in case of necessity, allows the safety functions to be maintained and pre-
ventive operations to be carried out on the plant during the whole. reactor operating period,
and the installation of the main part of the equipment in different compartments makes it pos-
Translated from Atomnaya Energiya, Vol. 54, No. 1, pp. 27-36, January, 1983. Original
article submitted July 6, 1982.
0038-531X/83/5401- 0019$07.50 ? 1983 Plenum Publishing Corporation 19
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
cA U
A Z
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
sible to ensure monitoring or the reactor in the case or an emergency situation (e.g., tire)
in one of the compartments.
The range of measurements of the neutron flux density from thesubcrtical state of the
reactor up to 120% reactor power amounts to 10-12 decimal orders of magnitude. For this the
flux density in the IC for the VVER-440 varies within the limits 10-1 to 1.2.1010 neutrons/
i
(cm2?sec), and for the VVER-1000 within the limits 10-1 to 6.108 neutrons/(cm2?sec). The
whole range is divided arbitrarily into three subr.anges: source (SR), intermediate (IR),and
power (PR). For reliability of carrying out the safety functions, overlapping measurement
subranges are provided, from 1 to 1.5 decimal orders of magnitude.
A type KNK-15 fission chamber, operating in the pulsed regime, is used in the source
range; a type KNK-4 helium ionization chamber is used in the intermediate range; and a type
KNK-3 boron ionization chamber, operating in the current regime, is used in the power range.
Each chamber operates in the current regime. The ionization is enclosed i~n the pressure ves-
sel and is shielded with electromagnetic screens, insulated from the chamber and the outer
vessel.
In order to Increase the operating lifetime, provision is made for moving the IR and SR
detector units by transfer mechanisms (.TM IC) with automatic anal manual control, a:nd for a
device for monitoring the position of the detector units (Fig. 2). The location of the de-
tector units (DU) is shown on the unified control panel (UCP). The system of control with
the transfer mechanisms automatically withdraws the detector units from the zone of substan-
tial neutron fluxes during bringing of the reactor to power; and ensures their emergency n-
sertion in the operating position in the case of transmission of a scram signal (SS) or a.
reduction of the reactor power. The manual control units provided for transfer of the detec-
for units have priority over the automatic control. The power range detector units are not
moved during reactor operation; their radiation stability is ensured by the corresponding
parameters of the IC, and by the choice of structural materials- and the communication line
cable. In order to increase the noise-proofing of the data-measurement channels, the signals
from the detector units are amplified and shaped in amplifier and conversion units (ACU), lo-
cated near the outlet from the IC channels of the detector unit communication cable.
The signals, shaped according to duration and amplitude, are relayed-along cables with.
a length of 150e coupling system ensures electrical coupling of the FPU with the detector units, and
also contains the pulse counters and timer devices, essential fpr the input of data from the
DU and for time tracking. The central processor (CP) processes the data, and controls the
operation of individual systems of the unit and the interchange of data; the PM, with a cap-
acity of -2 kbytes, serves for the storage of the FPU operating programs; and the OM, with a
capacity of 256 eight-discharging layers, serves for the storage of the intermediate results
of processing. The analog signal extraction equipment converts codes into voltage and feeds
the signals to the pen recorder, RPL, APC, and computer. The digital signal extraction equip-
ment provides the delivery of data in digital form to the presentation system, and also issues
signals which exceed the threshold settings. The monitoring, indicator, and manual control
unit effects the control by automatic monitoring of the soundness of the FPU, indication of
the quantities being measured, the admission and transmission to the central processor of the
values of the settings from the control desk, and also the manual control of the different
operating regimes of the FPU, necessary for carrying out startup adjustment, preventive, and
maintenance operations. Regimes for the automatic monitoring of serviceability, such as mon-
itoring the serviceability of the detector units by the presence in the FPU of an input fre-.
quency, and for continuous and test monitoring, are also provided for in the FPU.
The constitution of the complex was refined by taking account of the long-term structure
of the control and safety system, which provides for the presence. of three safety systems (Fig.
7).
The NFME complex contains three subassemblies: The first and second main subassemblies
comprise three neutron flux measurement channels over the whole range of variation, including
Analog Digital
outputs output
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
I G
?r-I O , r-I
m I ?~ ? ^ O
D, G .i..i H F+
,L O U ~, O ~
~ ~ ro o
~, oL? w p u u
~ G ?~
?c7 N co +~. G G
?rl G~ ?rl ?ri. ?rl
~. rl n1 C ~..
tii Sa. C- r-I U
~ ai ro
~ O G N ro'
~ G.:~. ,~ a aG
+~ ..L H G
(fi ~ F+ ?r-I r-I ?i-I. .
~, ;~ o u.i ro nc
pAOa
U
O C. D 41 b-~, r-
t~ G pa G P+ U1
.~: w Fn ,L
G ?~+ b +~
O O G
~+ ro ~ w
.,..I, .;.~, O
6C N G ~?+ G
v ~ ~ G o
s~ ?~ o ro ro
ro v now ~ u
.G ao o ro a ?~
U C ?-I C !A b
Ql ro ro ?rl G
~ ~
.~ G1. F-] b U
~ G ~ ~ ?~
~ 3 A H ~
o ~ AC F+ ~ W O to
I~t' I ~ ~\ I I R~ ^ o va~i o00 ~ ~ ICI (n J-J c E-~ ~ O
~ .C R'+ 7 'rl rv 1.r
E V] O' C ~ d
~~~
cn ?~ ?~I ?~I cn ,n
C k Cb?.i~~
O 4l G G C b Sa
+~ ;n oD 4l cn
C ro .?. G U 7 ?rl~
N v ~ .~ .r{ ?r?a 'b
O [n m ro
tr ? ^ v G s~ a
in u o v Q
~ ro~ o u a
- v s~ ro o
ao m ~ +~ .~ ro
?~ ?~ v ~ a ca m
w u~ ro o o v
\ ~~ ~ N N G Q O
~ N ~ [a o0 00 .r
m r. n .^ n
x ac ro.
LJ__I ~~ ~ py ~ Z, y G }-i
G ui 4 ~ ?~. u
H G O O U~
H ?~ ?+ N O
r-i >a a a s~
t O ?^
H ~ b .-I G t~
U G ro N 'ri ~,
? v ro G a. G ro
w +~ o a ~
w v?1 ov0 ~ ~ aD ai
Z ro u p, G ?~
v ~ G ,,~ v
00 0 0 ? ^ +~
C u w ~, u N
~ ro ~ ro v ~
~ sa v ~ m ?;~
a~ ro oo a +~
E OD ~ G rn +~ ro
aGroro~~GS~
> ?~ ?,~ v ~ ~o v v
o ~x e a
v o w a ro o0
~~ G~fa ro oomo
?~ ~ ~ ?~ ro
~ s~ ^ ro b . ^
O. cA r-I
~?, ~ ~ ~ .-. ~ .-. N
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
the ilrs~ scdLCUY equipment, ~r~n~ ana Lne measurement cnannels or Lne xuel recnarging monitor-
ing system (FRMS); the FSE and FRMS channels are not involved in the third subassembly - mon-
itoring of the reactor in the source range (SR), intermediate range (IR) and power range (PR)
is effected by means of this subassembly. The measurement channel of each subassembly con-
tains: a tandem wide-range detector unit, installed in one ionization chamber channel without
transfer during operation, and an amplifier conversion unit (ACU). (one to each ionization
chamber); a DSPE rack, including functional processing units and a logic unit; and an analog
and digital data presentation equipment, and an assignment setting unit (ASU) with respect to
power.
From the first and second subassemblies of the NFME issue three discrete scram signals
(SS) according to "2 out of 3" logic, three discrete threshold safety signals (.TS) to the con-
trol and safety panel room, and analog signals for the control and dumping of the reactor pow-
er (by the RPL and.APC).
The power and period measurement channel data arrive at the digital indicators and pen-
recorder potentiometers selectively through any measurement channel. The data presentation
units from the three subassemblies are duplicated at both the UCP and at the RCP.
Centralized data extraction on a CRT screen is provided for in analog (histogram) and
digital form from. the first and second subassemblies.
The threshold settings with respect to the period are assigned in channel (within the
DSPE), and the settings with respect to power in .the startup-range (_SR and IR), are centralized
from the assignment setting units in each "triplet"; in the PR the power settings are assigned
individually.
The design of the measurement channels, by varying the composition of the basic equipment,
allows conversion to shaping the warning signals by ''2 out of 4''' logic. Cgrivers.on to ''''2 out
of 4" logic can be effected easily after introducing variations in the structure of the nu-
clear power station safety system, associated with the provision of. a power supply system
with cable cuttings and runs, and with the acceptance of the NFME signals by other systems.
Experience in the operation and commercial manufacture of third-generation NFME confirms
the long-term prospects of the accepted constitution of the structure and the competence of the
functions performed, in order to ensure the safety of nuclear power station reactors. The
flexible and all-,purpose structure of the data-measurement channels, by a simple means, allows
the significant dif~erence in the neutron flux monitoring ranges and operating conditions of
the VVER-440 and VVER-1000 to be taken into account, and.also allows the neutron flux in re-
search reactors and critical assemblies to be monitored.
The high parameters concerning the reliability and metrological qualities have been con-
firmed, ensuring monitoring and control of the reactor power by the NFME signals during the
running time, with sporadic calibration by the value of the thermal power.
Subsequent improvement of the structure and elevation of the technical parameters of the
measurement channels are based on the extension. of the range of application of the standard
IC, the cables, and the use of digital processing methods. The achievement of new technical
solutions will ensure elevated parameters of accuracy, stability, reliability, and noise sup-
pression, and will reduce the power requirements of the equipment.
The combination of new technical solutions will make it possible to considerably simplify
the layout of the reactor, and. to increase the efficiency, reliability, and safety- of nuclear
power stations, taking into consideration possible extreme situations.
The authors express sincere thanks for valuable advice and constant interest in the paper
to V. V. Matveev, and. also to the following group of colleagues for having participated in the
development of the overall concept of the construction of the NFME complexes. and in the solu-
tion of a number of technical problems, and for having rendered assistance in introducing the
NFME complexes into operation: T.V. Andronova, V. A. yosnesenskii, A. N. Kamyshan, A., V. Ku-
priyanov, V. F. Lomzin, K. I. Lyubetskii, G. L. M.ikhailov, V. I. Mukhin, Z. V. Sokolov, and
others.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
1. V. A. Sidorenko, Problems of the Operating Safety of VVER Reactors [in Russian], Atomiz-
dat, Moscow (1977).
2. Regulations for the Safety of Nuclear Power Stations IBYa-04-74 [in Russian], Atomizdat,
Moscow (1976).
3. Recommendations of the International Electrotechnical Commission. General Principles
Relating to Equipment for Nuclear Reacttrs [in Russian], Publications 231, 232, and 323,
Geneva (1967).
4. I. E. Burenko et al., in: Proceedings of a Symposium of Specialists of CMEA Member-
Countries on "Strucutre of Systems of Equipment of Nuclear Instrument Manufacture" [in
Russian], Moscow (1976), p. 13.
5. System of Unified Standard Designs of Aggregation Complexes-for Hydroelectric Power. Sta-
tions. Types and Basic Dimensions [in Russian], GOST 20504-81.
6. A. B. Dmitriev and E. K. Malyshev, Neutron Ionization Chambers for Reactor Technology
[in Russian], Atomizdat, Moscow (1975).
INDUCED ACTIVITY OF CERTAIN CONCRETES IRRADIATED AT A 680-MeV PROTON
ACCELERATOR
V.
F.
Kas'yanov, A. N. Kargin,
M.
M.
Komochkov, B. V. Man'ko,
and
B.
S. Sychev
One method of decreasing the induced activity at present-day accelerators is to use build-
ing and structural materials which are not highly activated in particular concrete. Nachtigall
and Charalambus [1] investigated, in detail, the laws of formation of 24Na in the concrete at
the CERN accelerator. Estimates of the activation of the concrete shield of a proton accel-
erator reported in [2] showed that 24Na made the largest contribution to the total dose rate
from the induced activity. Experimental studies of the activation of concrete at the Berke-
ley betatron [3] led to the same conclusion.
Study of the induced activity of building and structural-materials, including ordinary
concrete with a density of 2300 kg/m3 [4, 5], showed that the principal radionuclides are
24Tda 22Na 56T,1n 54Mn and 'Be and that of these 24Na and 22Na
, , , , , present the greatest radia-
tion hazard. It was shown in [6] that the induced activity of lime concrete was 3-10 times
lower than that of granite concrete.
We present calculated and experimental values of the induced activity of concretes of
various cherical compositions when irradiated in a field of scattered neutrons in the 680
MeV proton accelerator room. The flux of scattered neutrons at the places where the samples
were irradiated at the JINR synchrocyclorton was measured with activation detectors [4]. The
values of the neutron flux in various energy groups were used in the activation calculations.
The technological and chemical compositions of the concretes chosen for the study are
listed in Table 1. Ordinary concrete with a density of 2300 kg/m3, the most commonly used
shielding and structural material at accelerators, is taken for comparison. The chemical
composition of Portland cement, the principal binder in concretes, .is listed below by wt.
[7]:
Na
0,1 Al
3,05
K
1,1 Fe
2,37
Si
11,45 0
36,39
Ca
39>2 Other
0,68
Mg
3,01
S
1,33 Loss in calcination.
1,32
Translated from Atomnaya Energiya, Vol. 54, No, 1, pp. 36-38, January, 1983. Original
article submitted Decetnber 15, 1981.
0038-531X/83/5401- 0031$07.50
? 1983 Plenum Publishing Corporation 31
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approve C 021968000300020001-9
d For Release 2013/02/06: IA-RDP10-
O O
a
o ~
o
0
0
_
"~"
~
~
c
~
G
H
N :D
:~ ~
~
m
O
0
0
N cD
~ O
O
N
T
O
O
~
~ m
~^~ 00
C.
a0
-~
U
O n
W
~
O O
N
[~
~Y'
N
T
~
O ~A
~J ~
O
~
:D
~T
[.
N O
^L N
L~
l~
G7
m
G
00
Q1 :D
C+:I
N
m ~
o
0
~'
m
?
m
c
i o
c
~
o c
o
0
?p
~ m
o~0
c~i
0 0
0
0
~
m
O t~
m
N
o0
U
W e-i
~
m
N
m
~ o`no
~
n
n'
N m
c i
o
x
m
co
cG
N
c~
O
N N
O
O
z
o
r
ti
o
0
i ~
on
c
o
`~ x M
Q ~ ~
~ N
m
o
a
j
I N ~'
"i
I
~,
O u
~
.'~
^~
c
~
a
N
~
~
V
v
~
i
N
ti ff
1
1 T
~ m~m~~ C I
~ Ta
~
b
~
~
~ ~
~
~ ro
'
FS
o.
n.
a~
N ~
a~
N
~ ~
N N
~-' ~yy
i 0
N ~
~
~ N
i~
O 'C3 U O .D N
'
'
'
~
o0 -O ~
~i
O b ~
N
~+
N N
~ y.
'
N N
O N m
Oy
~
~
x,~
w?rl
U bQ
~
~
v U
.U.
~ ~
rU.
CU"
C
~
~
?
"
?
m ~v
O
~
`3'd
~
~~~
~~
r
a
~i bA
N H
~ N
O G
a~+
Oq
N
u7 G' p~ U N
O G' F.'
W N zNU.
c, N
L..
1
~
Q'O
p. U ~,
O U
C3.
Qb
v,.a
N N
~
O
~
O
U
q~~~~ C
G U S
~ U N
f+
O
~bqF, O ,y w O
~Fa.. bq U O bq U
;d E" O
r.~,ti U
r~i ~ N
O N U
~
O a
O
a
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
0
U
Fig. 1. Decay of induced activity of concretes after irradiation:
calculated, Q, ? , ~, ^) experiment; ? , 0) ordinary concrete.
with granite and limestone aggregates; O, ^ ) polymer concrete with
granite and limestone aggregates, respectively; ~T = 2.8.105 neu-
trons/cm2?sec; ~p 0.62.105; ~E=2-so.MeV = 3.0.105; ~g>z? MeV =
1.7.105.
In polymer concretes, Portland cement is replaced by an epoxide resin. Crushed lime-
stone is used because its chemical composition approximates that of marble (CaC03), which is
not highly activated [5, 8].
The specific induced activity Q of the concretes (mg-eq Ra/g) was calculated from the
expression
Q (t, i, n, k) = 8 4 3.714~~ ~ 403 ~ ~ ~ F 6~m> k~i) 1 e_~zT (1- a-~'zt ),
(]) ('m) i mk tl ~jn 7 f1t
where t and T are, respectively, the irradiation and cooling times,Zsec; Fmk,mneutron flux in
the m-th energy group in the k-th neutron distribution, neutrons/cm ?sec; oi? cross section
in mb (1 b = 10_.ze mZ) for the formation of the i-th radionuclide from the j-~h stable element
of atomic number A? when bombarded with neutrons of the m-th energy group (these cross sec-
tions were taken mainly from [9, 10]; n. weight content of the j-th element in the material
of the n-til composition, %; ~i, decay ct~nstant of the i-th radionuclide; k~, gamma constant
of the i-th radionuclide, (R/h)/(mCi/cm2) (1 R = 2.58.10-`' C/kg; l Ci = 3.700.1030 Bq). The
calculated values of the specific activity Q (t, T, n, k) are shown in Fig. 1.
Concrete samples cast in the.form of cubes 40 mm on an edge were irradiated for 180
days in the field of scattered neutrons in the synchrocyclotron room. The induced activity
was measured with a spectrometer with a 36 cm3 Ge(Li) detector and an AI-8000 pulse-height an-
alyzer. The spectrograms were processed on a P4insk-2 computer. Radionuclides were identified
by comparing the measured y energies with tabulated values [10, 11]. The absolute values of
the induced activity of the concretes were found from a calibration: silver nitrate (AgNOs)
of known specific activity was dissolved in a cube with an unsolidfied mass of concrete hav-
ing a density close to that of the concretes under study. The efficiency of recording Y.rays
was determined by measuring this sample on the spectrometer,
The measured and calculated values of the specific activity of the concretes studied are
compared in Fig, 1. The calculations took account of the contributions of the following radi-
onuclides identified in the irradiated samples: z?Na,. 22Na, 56Mn, 5?Mn, 52P4n, 'Be, `ZK, and
?3K, The total error in the measurements of the activity does not exceed 30%, taking account
of the error of the volume calibration.
The calculated and experimental results are in satisfactory agreement. The average var.-
fiance of these data is 1.5. The computational errors are a result of using inaccurate values
of the neutron flux, the reaction cross sections, and the chemical composition of the con-
cretes.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
1? auiiuiiaiy, l~ wa5 es~a~115uea LnaL Lne princepat raatonucttaes aeLermtntng the induced
activity of the concretes tested are the following: in the first 5 h after irradiation, 2`'Na
and 56Mn; for T = 5 h to 4 days, 2`'Na; for T > 4 days, 22Na, 54Mn, and 'Be. The main contribu-
tion to the induced activity of the concrete comes from Na, Mn, Fe, A1, and Si, listed in or-
der of decreasing contribution to the total specific activity.
The concretes studied are less highlyactivated than ordinary concrete. By replacing
Portland cement by epoxide resin, and using the same aggregates, the induced activity was de-
creased by a factor of 1.5-2 for T < 30 days. By replacing the coarse and fine aggregates
(crushed granite and quartz sand) by crushed limestone and sand, and retaining Portland cement
as the binder, the activity of the concrete was decreased by a factor of 3-8 for the same val-
ues of T. Finally, by replacing the Portland cement binder by epoxide resin, and the crushed
granite and quartz sand aggregates by crushed limestone and sand, the induced activity of the
concrete was 8-14 times smaller than that of ordinary concrete.
In conclusion, the authors thank V. M. Tsubko-Sitnikova, M. I. Fominykh, and E. T. Kon-
drata for making available the measuring complex and the Minsk-2 computer.
1. D. Nachtigall and S. Charalambus, CERN, 66-28, Geneva (1966).
2. T. Armstrong and J. Barish,Nucl. Sci. Eng., 38, 265 (,1969), ORNL-TM-2630 (1969).
3. K. Goebel et al., CERN, 71-21, Geneva (_1971).
4. V. F. Kas'yanpv et al., JIPdR 816-8899, Dubna (1976).
5. V. F. Kas'yanov et al., JINR 16-12/375, Dubna (1979).
6. V. F. Kas'yanov and P. A. Lavdanskii, At. Energ., 45, 123 (1978).
7. D. L. Broder et al., Concrete in Nuclear Reactor Shields [in Russian], Atomizdat, Moscow-
` (.1966) .
8. M. Barbier, Induced Radioactivity, North-Holland, Amsterdam-London (1969).
9. L. R. Kimel' and V. P. Mashkovich, Shielding Against Ionizing Radiations (Handbook) [in
Russian], Atomizdat, Moscow (1972).
10. V. S. Barashenkov and V. D. Toneev, The Interaction of High-Energy Particles and Atomic
Nuclei with Nuclei [in Russian], Atomizdat, Moscow (1972).
11. G. Erdmann and W. Soyka, Gamma Lines of Radionuclides, KFA (,1974).
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
RADIATION CONDITIONS IN A 16-MeV ELECTRON MICROTRON ACCELERATOR
A. G. Below, G. A. Komendantova,
Yu. G. Teterev, and A. P. Cherevatenko
Microtons [1] recently have been increasingly employed in science and technology. The
16-MeV microtron of the Joint Institute of Nuclear Research is used as the basic unit pf neu-
tron activation and Y activation analysis of geological samples [2] and has the following ba-
sic parameters:
Energy Eo of the accelerated electrons
Average current of the accelerated electrons
Repetition frequency of the radiation pulses
Duration of Radiation Pulses
Efficiency of outputting the electron beam
Slowing-down target
Beam power on the target
Neutron converter
Integral neutron yield
Neutron moderator - graphite assembly with dimensions
Volume of the installation site
Ventilation rate
Flow rate of the water cooling the microtron
16 MeV
up to 30 uA
50-800 Hz
2.5 usec
X90%
2 mm tantalum
up to 0.48 kW
Uranium + beryllium
3.1011 neutrons/sec
l x l x l m
140 m3
1400 m3/li
2 m3/h
The electron beam, accelerated to 16 MeV, is extracted with the aid of a magnetic channel
frpm th.e accelerator and directed with a rotating magnet either tpwards the center of a graph-
ite cube (where the tantalum bremsstrahlung target and, transverse to it, the natural uranium
neutron converter surrounded by beryllium are located) or to a channel at the end of which
there is only the bremsstrahlung target (Fig. 1). Samples to be studied with. neutron activa-
tipn are placed into the graphite cube used for thermalization of photoneutrons. The micrp-
tron is mounted in a room on the second floor level of the Laboratory Building. Both the
ceiling and the floor have a thickness of 1.6 m and are made of conventional concrete.
During the operation of the .equipment, the mcrptron components at which electron losses
pccur during the electron acceleration and the outputting from the microtron and the brems-
strahlung target proper are sources pf Y bremsstrahlung. Neutron sources are provided by the
neutron converter and by the components of both the microtron and the equipment which are sub-
jected to irradiation by y quanta of an energy exceeding the threshold of photonuclear reac-
tions (Ethr = 10.8 MeV in the case of copper). After switching off the accelerator, residual
radioactivity persists in the components and in parts of the microtron, the bremsstrahlung
target, and the converter. During the first few minutes after switching off the microtrpn,
the uranium converter emits delayed fission neutrons.
The bremsstrahlung from the heavy target has a sharply pronounced angular distribution
in a solid angle of about 15? in the direction pf electron motion (at Eo = 16 MeV) [3]. The
angular distribution of the photoneutrons from the uranium converter can be considered isotrop-
ic [4], whereas the energy distribution resembles the fission spectrum with an average energy
of ~1.2 MeV [5].
The goal of the present work was to obtain empirical information on the factors epntribut~
ing to the radiation hazard anal on the sources in the unit during its operatipn, and to clar-
ify the possible use of the information in industrial equipment for the purpose of regular
monitoring.
Translated from Atomnaya ~nergiya, Vol. 54, No, 1, pp. 38-43, January, 1983. Original.
article submitted March 22, 1982.
0038-531X/83/5401- 0035$07.50
? 1983 Plenum Publishing Corporation 35
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
iiv/ti~V t'll.lvlJ C11YL 1'1Lu li1V LJ VL LVtlU iH11V1V L'1VLV 11 ViC 11VV
The high pulse duty factor (1000-8000) and the rather high limit energy of they quantum
spectrum before and behind the shield result in conditions of measurement for which.-most of
the standard instruments of radiation monitoring are not designed Therefore, it was Y:nves-
tigated whether the readings of the dosmetry instruments are adequate under the conditions
of microtron radiation, and the operational limits of the instruments were determined.
LiF TLD-700 thermoluminescence detectors of the Harshaw company were used as the basic
dosimeter of the Y radiation [6, 7]. The TLD readings are not affected by~ the fact that the
radiation is pulsed, and their dependence upon the radiation hardness does not exceed 20% in
the case of bremsstrahlung with an effective energy of up to 10 MeV [8]. I't .s assumed-that
behind the shield and beyond the limits of the direct bremsstrahlung beam, equilibrium. exists
between the charged and the neutral components. Figure 2 shows the results of a comparison of
TLD-700 readings with the readings of other y radiation dosimeters: dosimeters making use of
the RUP-1 gas discharge counter [9] or the DRG3-D1, DRG3-03, and DRG3-04 scintillation detec-
tors [10]. Th.e scintillation dosimeters measure the bremsstrahlung dose rate adequately in
the range P G 500 uR?sec-1 (1 R = 2.5810-'` C/kg). It was a surprising finding that even
the RUP-1 i~istrument with a Y detector having a dead time which is comparable to the pulse
length of the microtron radiation renders correct readings at P < 40 UR~sec-1. Tfiis fact
seems to be associated with the low efficiency of y quanta recording (at Py ~ 40 uR?sec"~',
the RUP-1 substantially reduces the dose rate which can be measured).
A noticeable sensitivity to y bresmsstrahlung of the microtron at P ~ Q,5 uR?sec-'1 of
the industrial DNA-1 neutron dosimeter [10] and the RUPrl radiometer with neutron detector
was established. In measurements behind the neutron shield, where, according to the results
of measurements with nuclear emulsions [11] and of REM-2 recombination dosimeter measurements
[12], there should be practically no neutrons (the measured coefficient of the radiation qua.1-
ity rendered the value KK ~ 1), the RUP-1 instrument with neutron detectors recorded about
40,000 neutrons/(cmZ?sec), whereas the DNA-1 instrument rendered a value of about 200 mbohr?
h-1 regardless of which point was used in the measurements (1 bolir = 1 cSw~). This result,
obtained with the neutron instrument equipped with scintillation counters, can evidently ex-
plain the fact that the signals produced by powerful high~nergy y quantum ra.diatioc~ pulses
exceed the discrimination threshold of the electronic circuits of those instruments.
The SNM-14 corona counters of slow neutrons are insensitive to nonpulsed Y? radiation
fields at PY < 1500 R?h-1 [13]. The effect of pulsed bremsstrahlung of the microtron upon the
operation oY the counter placed at the center of a paraffin cylinder with a wall thickness
of 120 mm was investigated. Such a detector is considered a dosimeter [14, 15]..
The pulses obtained from the neutron detector were recorded with a scaler and, at the
same time, observed on the screen of an oscilloscope which was synchronized with the trigger
frequency of the microtron. Since the "lifetime" of neutrons before their recording is, in
the paraffin cyliude~, much longer than the duration of the bremsstrahlung pulse, the neutron
pulses on the oscilloscope screen are statistically distributed, whereas the y rada.tpn pals-
es are synchronized with the frequency of microtron .operation, The neutron detector was mount-?
ed in th.e hall of the microtron when the electron beam was: extracted from the microtron and
directed onto a slowing-down target surrounded by a lead trap. By varying the thieknes?s: of
the lead walls., the flux ratio of Y quanta and neutrons could be changed. The P~ value was
measured with. a DRG3-04 instrument. Up to an average counting rate of 10` sec-1 the neutron
detector readings were linearly proportional to the electron current directed onto the brem-
sstrahlung target. At PY 500 uR~sec-1 on the scope screen. there appeared a synchronizing
pulse originaing from the bremsstrahlung pulse, but the pulse amplitude exceeded the level
of the corona noise only by a factor of 2; the s~nchrgni;zing pulse was still discriminated by
the recording system having the threshold sensitivity recommended in the specification of the
SNM-14 counter. In the case of microtron operation with a neutron converter in a graphite
cube, nonlinearity resulting from counting losses was observed at counting rates in excess
of 2.5.10? sec-1.
An SNM-14 counter combined with a polyethylene moderator ('wa.ll thickness 124 mrn and 0.6
mm cadmium layer at a depth of 55 mm in the moderating layer) was. used as an active neutron
dosimeter [16], Indium-activation detectors. in a paraffin sphere wth.a diameter of 280 mm
[17] or in a polyethylene sphere with a diameter of 2.54 m03
P~ TLD-'700, pft/sec
Fig. 1 Fig. 2
Fig. 1. Plan of the microtron site: a) high-frequency generator; b) microtron; c)
outputting device with shield; d) switching magnet; e) graphite tube; f) y channel;
g) 20-cm-thick lead trap of the y channel; h) roll door of brass (the figures denote
the number of the point at which the radiation conditions were studied).
Fig. 2. Dosimeter readings in the microtron radiation field for instruments DRG3-
O1, DRG3-03, and DRG3-04 (~), and for RUP-1 with gamma detector (O), compared with
readings for TLD-700.
~ 9~y [~'i~~luwiuuwllL~yi v
? >Oo
04
~ SO
X
E~, MeV
~~~ >0' i0f ~ ~~ 700
c?i MeV Eo, MeV
Fig. 3 Fig. 4
Fig. 3. Typical ratios of the equivalent dose rates H for various types of second-
ary radiation before the shield of electron accelerators of energy Eo.: 1, 2) brems-.
strahlung occurring under the angles 0? and 90? relative. to the direction of the
electron beam; 3) neutrons; 4) induced radioactivity; 5) muons (only under the angle
0?).
Fig. 4. Thickness values for tenfold attenuation in dependence upon the energy Eo
of accelerated electrons and the energy En of the neutrons: 1) conventional con.-
Crete; 2) steel; 3) lead.
The readings of the neutron dosimeters in the microtron radiation fields were verified
by comparing with an equivalent dose determined from the energy spectrum of the neutrons.
The neutron spectrum was restored by indium activation or with a TLD 600/.700 pair in spheri-
cal polyethylene moderators of various diameters [19] and subsequent data processing with a
program using the statistical regularization technique of [20]. Comparisons were made for
microtron operation with a slowing-down target comprising a neutron converter and the graphite
cube (mode 1), as well as for microtron operation with a bremsstrahlung target of the y chan-
nel (mode 2). Neutrons with an energy below 1 keV dominate (~90%) in the spectrum of mode
1, whereas the spectrum is enriched by fast neutrons in mode 2. The readings were compared
for all neutron detectors employed at points 7, 9, and 10 (Fig. 1); at points 3, 4, S, and 6
the comparisons were made for detectors in the polyethylene sphere with the diameter of 254
mm.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
i~ waa vuacivCU utGL LL1C JLV 17-14 ueuLron counter in the parauin moaera.tor with a wall
thickness of 120 mm renders values which exceed the most probable dose by a factor of up to
10 in mode 1 and of up to 3 in mode 2. The SNM-14 counter in the combined polyethylene mod-
erator renders values exceeding the most probable dose by a factor of up to 3 in mode 1 and
at point 2 behind the rolling brass door in mode 2, whereas at the other points the readings
of this counter in mode 2 coincide within the error limits with the most probable dose value.
When indium is used in the paraffin sphere with a diameter of 280 mm or ~n the polyetliyl-
ene sphere with the diameter 254 mm, the values are too large by a factor of 2-2.5 in both
modes of operation. The same value is observed with the TLD 600/700 pair in the polyethylene
sphere with the diameter of 2.54 mm. Abnormally large readings of a dosimeter consisting of
an SNM-14 counter in a cylindrical paraffin moderator seem to be related to the moderator de-
sign in the case of a "soft" neutron spectrum (the moderator comprises a transverse channel
running through the center to accommodate the SNM-14 counter; therefore, the counter can be
hit by slow neutrons which did not pass through the moderator).
RADIATION CONDITIONS
External Radiation. Bremsstrahlung is the main component in the radiation field of elec-
tron accelerators. This situation is illustrated in Fig. 3, which shows qualitatively typical
ratios of equivalent dose rates normalized per unit power of the primary electron beam f.or
various types of secondary radiation generated at a distance of 1 m from an unshielded target
which is bombarded by electrons with the energy Eo [21]. The width of the shaded area is
affected by quantities such as the target material and the target thickness.
Figure 4 displays the length values for a tenfold attenuation of the equivalent dose in
abroad bremsstrahlung beam for the cases of concrete, steel, and lead; the values refer to
the angle 0? relative to the direction of the primary beam of electrons with the energy- Eoand
are shown in dependence upon the energy of the electrons incident upon the heavy target [21].
Figure 4 also shows the length values-for tenfold attenuation of the equivalent dose of neu-
trons in dependence upon the neutron energy En [22]. Obviously, in the case of concrete the
length of the tenfold attenuation of the equivalent dose of bremsstrahlung generated by elec-
trons with the energy Eo = 16 MeV is about 100 g?cm 2, whereas in the case of photoneut.rons
with En = 1 MeV (average energy of the spectrum) the corresponding value is ~50 g?cm 2, i.e.,
the equivalent dose behind a concrete shield, as well as before the shield, results mainly
from bremsstrahlung. The inverse situation is observed in the case of iron (steel, brass):
the layer of tenfold attenuation of the equivalent dose of bremsstrahlung amounts to ~85 g.
cm 2 and to about T5O g?cm s for 1 MeV neutrons; the irradiation conditions behind the shield
are given only by the neutrons, not by the bremsstrahlung. Since the source of the secondary
radiation of the electrons is local (slowing-down target-converter), a local shield, e.g.,
of lead, is conveniently employed for reducing the bremsstrahlung level in the hall; in this
fashion the requirements to be satisfied by the main (concrete) shield can be reduced.
Local regions of bremsstrahlung unrelated to the target-converter were detected in the
initial period of using the microtron as a radiation source. At an overall background level
of ~0.1 uR?sec-1 the value P = 15 UR?sec-1 was observed at the control panel (point 10; see
Fig. 1) located opposite theYmicrotron resonator. When the coil which adjusts the electron
orbits and which is located near the resonator is poorly tuned, the P,~ value at point 10
increases up to 30 times. The distribution of the y radiation dose rate behind the shield
up to the wall which is perpendicular to the direction of the extracted beam was characterized
by two peaks, which attests to still another local bremsstrahlung source besides the brems-
strahlung target. In order to establish this source, TLD dosimeters were exposed in the hall
of the microtron (at 35 points) while the bremsstrahlung target was kept surrounded by a lead
shield. As a result, a source was detected at the transition from the magnetic channel to
the electron tube from which the beam with the final acceleration energy is extracted. In
order to improve the radiation conditions, a local shield was inserted for the bremsstrahlung
sources established: a 10 cm lead shield was placed inside the armored core of the microtron
magnet opposite the resonator; the region of the transition from the magnetic channel to the
electron tube was also protected by 10-cm-thick lead. Ten cm lead (along the path of the
electrons) was inserted inside the graphite cube behind the slowing target-converter. This
did not influence the distribution of both the thermal neutrons and the resonance neutrons
inside the cube. The inner wall of the cube was laid out with 5-cm-thick lead bricks on a
plane perpendicular to the beam axis. The wall portion which is perpendicular to the electron
beam extracted toward the graphite unit e (.see Fig. 1) was reinforced by 40 cm conventional
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
1t1DLL~ 1, iveu~rvn ana y Juan r.a Lose xat:es
at Various Points of the Microtron Hall
Point of mea- I Mode cf I Neutrons, bohr ? I y radiation,
surement (see opera h-t R ? h-t
Fig, 11 ~ lion
1
1
20110
1000
2
1
2010
360
3
1
6,1?3,4
24
4
1
5,3?3,0
380
5
1
'L,6?1,G
77
6
2
17,711,2
21
7
1
0,710,4
G
2
3,910,4
18
8
1
-
11.10-3
9
1
2,5.10-3
2.10-3
2
(13,32,9)?90-3
`L,8.10-3
10
1
U,11.10-3
0,4.10-3
2
(0,58-U, 2)?10-3
0,7.10-3
11
1
1,0.10-3
1,8.1U-3
2
(5,4 11),5)?10-3
1.,4.10-3
TABLE 2, y Radiation Dose Rate at a Dis-
tance R from the Source Resulting from the
Induced Activity of Microtron Elements,
uR?sec-1
Time a
ter stopp
On the
surface
lowing-
dewntan-
Near the
magnet
Uranium
t
Reso-
ator
ing the
mtcro-
of the
graphite
talum
target(R
oke
above the
er
conver
(R = 50
n
(R = 30
tron
cube
5 emi
resonator
cml
cml
10 min
50
2,2
1,2
120
1,2
3p min
36
0,9
1,0
-
0,6
1,5 h
15
0,6
1,0
-
0,25
10 h
2,2
0,4
< 0,1
20
0,1
16 h
1,4
0,3
< 0,1
-
< 0,1
53 h
0,8
0,2
< 0,1
1,6
< 0,1
23 days
< 0,1
< 0,1
< 0,1
1,6
< 0,9.
concrete (i.e., laid out with blocks). Therefore, when the microtron was operated in mode 1,
the y radiation field behind the shield was practically homogeneous and the Py value decreased
by a factor of about 50 relative to the maximum values. In operation in mode 2, the brems-
strahlung target was protected by 20 cm lead disposed in the direction of the beam and. by 15
cm lead on the sides.
The radiation levels at the points of monitoring are listed in Table l for the case of
nominal operation of the microtron with an additional shield, The neutron dose was estimated
from the data of the neutron spectrum obtained with the multisphere technique of [19, 20].
They radiation dose in the hall of the microtron was measured with a TLD-700.
The integral neutron dose spectra observed at certain typical points of the microtron
are shown in Fig. 5. The ''softest" neutron spectrum corresponds to operation in mode 1. The
neutron dose at points 9, 10, and 11 (see Table 1) was measured with a 10 inch moderator and
increased liy a factor of two to obtain the probable dose,
');he "hardest" neutron spectrum is observed in mode 2 in the hall of the microtron. Be-
hind the shielding, the spectrum appears softened, but it is harder in a direction close to
the direction of the primary electron beam. The spectra at other points are of a type inter-
mediate with respect to the curves shown in Fig, 5. The dose rate of the neutrons and of the
y radiation near the building of the microtron amounted to 0.1-0.2 mbohr?h_,..
It follows from Table 1 that the neutron dose behind the rolling brass door (see point
9 in Fig. 1) exceeds the dose of the bremsstrahlung radiation. In such cases a shield or a
labyrinth of about 0.5-m-,thick conventional concrete should be placed before the entry.
Induced Activity. The dose rate resulting from the induced activity was measured in the
area of equipment elements at which the main electron losses occur, namely the slowing-down
tantalum target, the uranium~eryllium converter of neutrons, the resonator, and the output-
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
yd
0 0, 8
.a ?, 0, 6
~'
?C a 0,'4
U~0,2
0
. 70_p ~~-4 ~Q~
En, MeV
Fig. 5. Relative contribution of neutrons of energy En above
a given energy to the total dose at the microtron: 1) points 3
and 4 (see Fig. 1), mode l; 2) point 10, mode 2; 3) point 9,
mode 2; 4) point 6, mode 2.
ting unit; measurements were also made near the graphite cube at the point at which a person
stands while taking irradiated samples from the cube. It follows from the results of the
measurements (Table 2) that the dose near the accelerator is given mainly by the activity of
short-lived isotopes. .
With the microtron disassembled and the resonator removed, PY inside the brass chamber
and close to the walls amounted to 6-8 UR?sec-' 10-15 min after switching off the microtron,
but to only 0.3 uR?sec-1 after 17 days. Spectrometry performed with the aid of a Ge(Li) de-
tector on the removed resonator showed only the presence of the fi2Cu (T1/2 = 9.8 min) and
64Gll (T1/z = 12.8 h) isotopes. When the dose rate near the uranium converter was measured
10 min after switching off the microtron, it was estab lished that the main contributinn to
the short-lived component of the induced activity stems from z3eU radiation. Long-lived ac-
tivity appears basically in that part of the uranium converter into which the beam of brems-
strahlung quanta enters (in Table 2, the P value near this face of the uranium converter is
listed). The opposite part of the convert~r is characterized by an activity which is lower
by several orders of magnitude. The slowing-down target is replaced once every half year.
The values listed in Table 2 refer to the end of its service life. One minute after switch-
ing off the microtron the dose rate of delayed uranium neutrons near the graphite cube amounts
to about 3 mbohr?h'1, and it amounts to 0.2 mbohr?h-1 after 3 min.
Radioactivity of Air and Water. Activation of the air and the water was observed at
E? > 10.55 MeV at the nitrogen of air and at E? > 15.67 MeV at the oxygen of the water, re-
spectively. External radiation resulting from the radioactivity of the air and from the tubes
with water (cooling system of the accelerator) is significantly weaker than irradiation caused
by the induced activity of components and targets of the microtron. External radiation can
be disregarded as a radiation hazard.
During operation, the concentration of radioactive aerosols in the hall of the microtron
did not exceed 3.5.10-1fi Ci?liter 1 (1 Ci = 3.700.1010 Bq) in S activity and 1.7.10-16 Ci?
liter'1 in a.activity (background of the equipment). Inside the ventilation tube, at a dis-
tance of 3 m from the point of air intake from the hall, P~, < 50 >1R?h'1 immediately after
switching off the microtron. Assuming that exchange ventilation does not take place, the cal-
culated `'lAr concentration amounts to 4.10-10 Ci?liter-1, which is smaller than the admis-
sible concentration of this isotope in the air of places of work by two orders of magnitude
[23].
The value PY < 50 uR?h-1 was observed near the tube with the water for microtron cooling.
A spectrometric analysis of the isotopes precipitated by iron hydroxide from 5 liters of water
at pH = 8 was performed f_or 12 hours two days after taking a sample; measurements with a Ge-
(Li) detector with a volume of 40 cm3 did not reveal a measureable residual activity. Calcula-
tions made with the method of [24] have shown that the average annual concentration of radio-
nuclides in the water cooling the microtron is below the admissible concentration of open
water reservoirs when continuous operation of the microtron is assumed.
It could be concluded from our data that bremsstrahlung is the basic factor of radiation
hazard during operation of a 16 MeV microtron. Although the dose rate behind the shield of
the microtron hall can be significantly reduced with the aid of a local shield, the concrete
walls should have a thickness of not less than 1.5 m. When a roll door of steel or brass is
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
employ cu, a ..via~ic~.c Zvi ~viyc~iiyicuc~ iavyiiu~ii aiivulu vc Yivviucu civic ~iic uvvi vu a~2
side of the hall. For the energies and power ratings of the electron beam indicated above,
the level of the induced dose rate does not call for special measures in servicing, preventive
maintenance, and repair work on the microtron although the induced activity of the uranium
converter is rather high. The water cooling and ventilation systems of tfie microtron do not
require any special elements for obtaining radiation safety, but considerable air exchange
is necessary because a large amount of ozone is formed. Operational dosimetric monitoring
of the radiation conditions on the microtron should be made for the Y radiation of the instru-
ment with the DRG3-04 (or DRG3-O1 and DRG3-03) instruments and for the neutrons with an in-
strument using a corona counter of neutrons together .with a polyethylene moderator.
LITERATURE CITED
1. S. P. Kapitsa and V. N. Milekhin, The Microtron [in Russian], Nauka, Moscow (1969).
2. V. Ya. Vyropaev,JINR 14-9446, Dubna (1976).
3. Yu. N. Burmistenko et al., in: Accelerators [in Russian], Atomizdat, No. 10, Moscow
(1968), p. 238.
4. V. N. Tsovbun, JIPdR 16-7104, Dubna (,1973).
5. V. P. Kovalev, Secondary Radiation of Electron Accelerators [in Russian], Atomizdat,
Moscow (1979).
6. K. Barker, Solid State Dosimetry, CRC Press, Ohio (,1973).
7. M. Franz and V. Stolz, Solid-State Dosimetry of Ionizing Radiation jRussian translation],
Atomizdat, Moscow (.1973).
8. M. Ehrlich and C. Soares, NBS Technical Note 1119, Washington (.1980).
9. B. N. Nemirovskii et al., Radiation Meters jin Russian], Atomizdat, Moscow (1972).
10. V. F. Kozlov, Handbook on Radiation Safety [in Russian], Atomizdat, Moscow (1977), p.
321.
11. M. M. Komochkov and M. I. Salatskaya, JINR 816-8175, Dubna (1974).
12. M. Zel'chnskii and K. Zharnovetskii, in: Neutron Monitoring, IAEA, Vienna (1967), p.
125.
13. L. S. Horn (corn) and B. I. Khazanov, Selective Radiation Meters` jin Russian], Atomiz-
dat, Moscow (1975).
14, De Pangher, J. Nucl. Instrum. Methods, 5, 61 (1959).
15. L. S. Zolin, JINR 2252, Dubna (.1965).
16. Radiation Safety. Quantities, Units, Methods, and Instruments jin Russian], Atomizdat,
Moscow (.1974).
17. V. E. Aleinikoy et al., JINR 816-6790, Dubna (1972).
18. J. Tuyn, in: Proc. 5th Int. Conf. on Luminescence Dosimetry, Sao Paolo (1977), p. 288.
19. V. E. Aleinikoy et al., JINR 816-9123 [in Russian], Dubna (_1975).
20. L.S. Turovtseva,,'''Solution of improperly stated inverse problems by statistical regular-
ization (OBR-23 Program)," Preprint of the Inst. of Appl. Math.., Acad. Science USSR
(1975).
21. tiJ. Swanson, Radiological Safety Aspects of the Operation of Electron Linear Accelerators,
Technical Reports Series No. 188, IAEA, Vienna (1979).
22. M. M. Komochkov, JINR 816-7335, Dubna (1973).
23. Radiation Safety Standards NRB-76 [in Russian], Atomizdat, Moscow (1978).
24. M. M. Komochkov and Yu. G. Teterev, At. Energ., 34, No. 1, 17 (1973).
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
R.
I.
Lyubtsev, V. I. Orlov,
V.
S.
Belykh, A. G. Evdokimov,
V.
N.
Voichishin, G. A. Akopov,
V.
Ya.
Mishin, B. I. Rogozev,
M.
K.
Abdulakhatov, and E. M. Rubtsov
The basic characteristics of the electromagnetic mass separator started
Khlopin Radium Institute are as follows:
up
at
the
V.
G.
Sector angle of the analyzing magnet, rad
2
Average radius of the ion trajectory, m
1.0
Distance from the foci to the effective boundary of the
field, m
2.0
Ion current from the source, mA
1-10
Dispersion (for 1% Dm/mo),.mm
20
Resolution of mass peaks at half-.height
1500
Relative enrichment
not less
than 250
The mass separator was created based on the principles proposed in [1] and to a certain
extent is analogous to the setup described in [2] and [3]. The difference lies in the fact
that the mass separator examined here is intended for separation of radioactive isotopes. This
determined the arrangement of the equipment and the remote control of its operation.
The working chamber, analyzing magnet, and vacuum system are situated in a special cham-
ber (Fig. 1), equipped for first class work with radioactive substances. The vacuum system
and the system for feeding the source of -ions and other units and monitoring the operational
regime of the setup are controlled from a control panel situated outside the zone where the
work with the radioactive substances is performed. In order to simplify the periodic decon-
tamination of the working chamber, a collapsible protective stainless steel liner is inserted
into it.
In order to supply power to the analyzing magnet, a precision tliyristor current source
is used, which has considerable advantages over the usual generators. The instability of the
current in the windings exciting the magnet in the case of prolonged operation does not ex-
ceed 0.02%. No devices are supplied at the source output for smoothing the current pulsations,
but for current pulsation in the working region equal to 1.5%, the pulsation of the magnetic
flux in the gap of the analyzing magnet does not exceed 10-3%. This effect is explained by
the smoothing action of the vortex currents in the massive magnetic circuit [4].
The setup is distinguished also by the fact that due to the interchangeable attachments
on the pole end pieces, it is possible to separate isotopes of both heavy and medium mass
(10-250 amu). Th.e necessity for interchangeable attachments arises due to the change in the
topography of the magnetic field in the gap of the analyzing magnet due to saturation of the
steel in the pole end pieces when the current in the exciting windings changes. For this
reason, in order to ensure focusing of the beam in the other mass range, it is necessary to
change ,the shape of the pole edges.
We tuned the magnetic field of the setup by determining the topography of the field of
a real magnet from the ion trajectories and correcting it [3]; the intensity of the field was
measured by a method based on nutation of the total magnetic moment of protons j5]. The fine
adjustment of the field was performed by the partial beam method [3] using a beam of lead ions.
Due .to the correction of the field, the resolution of the mass peaks at half-height reached
1500 with a quite good shape of the peaks. Measurement of the background between peaks showed
that with the accumulation of nuclides, the relative enrichment will be of the order of 300.
Translated from Atomnaya Energiya, Vol. 54, No. 1, pp. 43-46, January, 1983.. Original
article submitted February 22, 1982,
42 0038-531X/83/5401- 0042$07.50 ? 1983 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
ila viuCr_ Lv JLQU111GC LIIC 1CE', 1111C Vl QI.L LLIlILL1QL1V11 Vl 1JVLVlJCJ QIILL 1llLLCQJC L11C l~LLQ11Lv
of their separation, a two-electrode system for following the position of the beam was us?ed
in the setup [6].
An IR-la ion source, which is a universal double crucible source eJith an incandescent
cathode, is used. In order to accumulate isotopes, a PP-1 detector is used [2].
The mass separator examined was used to obtain enriched preparations of 234U and 119mSn.
SEPARATION OF URANIUM ISOTOPES
The starting substance for accumulating preparations of 234U was uranium tetrachloride
obtained by chlorinzation of U3O8 in the flow of carbon tetrachloride vapor. Purification was
performed by sublimation in a vacuum.
In connection with the high hygroscopicity of uranium tetrachloride, after each separa-
tion operation it was necessary to decontaminate the source and to change the cathode. After
contact with the atmosphere, a sharp decrease was observed in the electrical strength of the
interelectrode gaps, and a degradation of the operational parameters of the tantalum cathode.
When the source of ions was recharged with matter without removal from the working chamber
(when the source was located in a dry nitrogen atmosphere), there was no disruption in its
operation.
Stable and easily controlled operation of the ion source was achieved with a flow rate of
matter of 140-160 mg/h. Such a flow rate of uranium tetrachloride was determined by an open-.
ing with a diameter of 1.2 mm in the output diaphragm of the crucible, whose working tempera-
ture was 320?C.
We checked the separation parameters in experiments on initial preparations with differ-
ent content of 234U, 23sU, and 23gU. The accumulation was conducted in copper accumulating
boxes with slit dimensions 3 x 70 mm. Analysis of the preparations obtained showed that the
coefficient of relative enrichment of neighboring isotopes was 250-270.
The utilization factor of matter (UFM) reached the highest values for large loads- (7-9 g
of tetrachloride, i.e,, approximately 50 h of continuous operation) due to the relative short-
ening of the time for introducing the mass separator into the e-stablistied working regime.
The starting preparation for accumulating 234U was uranium with a content of 715 at.%
of 234U; the main impurity was z3sU. As a result of reprocessing, specimens of 234U were ob-
tained with isotopic purity up to 95.25%. The UFM for 23~tU reached 4.8%, and the average
value of the UFM for all reprocessing procedures was 4.4%.
Targets containing the preparations obtained were used in work on determining he nuclear-
physical characteristics of 234U [7, 8].
Tin is a very inconvenient element to use in universal ion sources. With its low melting
temperature, tin has a very high boiling temperature and cannot be used in an IR-1a ion source,
Tin chlorides and fluorides, on the contrary, have a very low boiling temperature. The most
appropriate compound is tin dichloride SnClZ; this is the compound chosen as the working sub-
stance.
The temperature at which the working pressure of tin vapor is created in the discharge
chamber lies below the limit of the optimum temperature for the ion source. In order to de-
crease the conductivity of the vapor supply line, two diaphragms were mounted in the crucible
with openings having diameter 0.8 mm. The substance was loaded into a quartz ampule, covered
by a graphite plug with an opening with diameter 0.8 mm. Under these conditions, at a cruci-
ble temperature of 180-210?C, the flow rate of SnC12 vapor was in the range 70-90 mg/h, with
satisfactory stability of the flow rate and controllability of the discharge. The ion current
from the source in this case was 3-4 mA.
The resolution of the mass peaks at the focus of the setup, as in the range of heavy
masses, is of the order of 1500; the UFM for separation of tin isotopes reached 7-10%. It
should be noted that the UFM depends on the quality of the tin chloride prepared, If other
tin compounds are present in the working substance, then this index decreases to 4-.7%, The
best indices were obtained with the use of a working compound purified by sublimation in a
vacuum.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
Fig. 1. Diagram of the arrangement of the equipment in the cham-
ber: 1) box; 2) ion detector; 3) analyzing magnet; 4) ion source;
5) vacuum working chamber; 6) zone with maximum concentration of
aerosols when the working chamber is opened (in the region of the
ion source).
Q
100
10
i ~ ~ ~ ?i
0 1 2 3 4 L,m
Fig. 2. Nature of the distribution of the surface contamination
of the liner along the working chamber with separation of uranium
(1) and tin (2) isotopes; the origin of the length of the working
chamber is taken as the supporting flange of the ion source; the
dashed line shows- the interpolation of the section of the curve
represented by the continuous line,
In order to accumulate 119mSn we used a mixture of isotopes with the following composi-
tion (at.%): 99.2 118Sn; 0.8 119Sn; 0.01 119mSn; ? 1 113Sn; ? 1 6OCo. The specific activity of
the starting preparation was 0.5 Ci/g (1 Ci = 3.700.1010 Bq). As a result of reprocessing, prep-
arations of 119mSn were obtained with specific activity up to 40 Ci/g.
The tin isotopes were accumulated in copper accumulating boxes; the size of the input
slit was 8 x 70 mm. The source operated in a stable steady-state regime for approximately
70% of the reprocessing time for a single charge. At the end of the separation process, in
order to maintain the current from the source at a constant level, the power delivered to the
crucible heater is gradually increased. Preparations of 119mSn obtained are intended for pre-
paring Mossbauer sources.
While performing radiationally dangerous work, involving breaking the hermetic seal of
the working chamber (maintenance and decontamination of the ion source and liner, removal of
accumulating boxes, etc.), we investigated the contamination of the air and the surfaces of
the enclosure of the working zone, as well as the distribution of the surface contamination
of the parts of the liner as a function of distance from the ion source (Fig. 2),
We analyzed the aerosol contamination of air at all stages of the isotope separation pro-
cess. The aerosol concentration in the zone where the work was performed when the working
vacuum chamber was opened increased with time in spite of the measures taken to reduce it:
creation of a small flow of air directed into the chamber and use of polyethylene films to
cover the ion sources and parts of the liner before they were transported. When evacuating
the liner in the decontamination box after completion of the urnium isotope separation process,
the aerosol concentrationin the air in the zone noted in Fig. 1 (position 6) constituted 40
mBq/liter, with total activity of the reprocessed uranium equal to 4.108 Bq.
The total activity of ttie reprocessed 119mSn constituted 1011 Bq (2.5 Ci), but the aero-
sol emission into the atmosphere in the room was not observed over the entire duration of the
work with this substance.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
lue dc~lvl~y ur sue surfaces in Lne enclosure or the working zone remained at the back-
ground level, with the exception of separate sections, where local contamination was noted up
to 400 kBq/mz while working with uranium and tin.
The data examined lead to the conclusion that it is possible to use in.tiie setup starting
preparations with significantly higher activity. With some changes in the equipment, it is
possible to enrich isotopes of such nuclides as plutonium, americium, and curium, as well as
the fission products 85Kr, 9oSr, 93Zr, lo6Ru, 13sCs, 13'Cs, 14taCe, for which, when it is ne-.
cessary to obtain these preparations of high isotopic purity, electromagnetic separation, is
apparently most efficient.
The nature of the distribution of the surface contamination of the liner in working with
uranium and tin is shown in Fig. 2. The activity of the surfaces is shown along the ordinate
axis in arbitrary units.
Thus, the S2 type electromagnetic mass separator permits separating radioactive isotopes
in the interval 100-250 amu. The activity of the starting preparations for separation can
reach 10 Ci and ,higher. The parameters of the separator achieved make it possible to separate
isotopes with relative enrichment not less than 250. Isotopic preparations of uranium enriched
in 234U and tin enriched in 119mSn have been obtained.
LITERATURE CITED
1. A. F. Malov, V. A. Suzdalev, and E. P. Fedoseev, Zh. Tekh. Fiz., 35, No. 5, 914 (1965).
2. Production of Isotopes [in Russian], Atomizdat, Moscow (.1973), p. 565.
3. A. F. Malov, V. A. Suzdalev, and E. P. Fedoseev, Prib. Tekh. Eksp., No. 2, 146 (1969)
4. L. A. Tikhomirov, "Investigation. and development of thyristor precision constant current
stabilizers for exciting magnetic fields," Author's Abstract of Candidate's Dissertation,
LETI, Leningrad (1972).
5. A. I. Zhernovoi, Yu. S. Egorov, and G. D. Latyshev, Prib. Tekh. Eksp., No. 5, 71 (1958).
6. M. K. Abdulakhatov, G. A. Akopov, and V. S. Belykh, Preprint, V. G. Klilopin Radium In-
stitute, RI-136, Leningrad .(1980).
7. A. M. Geidal'man et al'., in: Abstracts of Reports at the 30th All.-Union Conference on Nu-
clear Spectrometry and Nuclear Structure, Nauka, Leningrad (1980), p. 149.
8. A. M. Geidal'man et al., Izv. Akad. Nauk SSSR, 44, No. 5, 927 (1980).
DETERMINATION OF THE COEFFICIENT OF ,ISOTOPE SEPARATION IN CHEMICAL EXCHANGE
BY THE METHOD OF MULTISTAGE EXTRACTION
S. D. Moiseev, V. A. Samoilov, UDC 621.039.32.2
and Yu. I. Ostroushko
The method of multistage extraction of isotope mixtures [1-3] is widely used for deter-
mining the coefficient of separation of the isotopes in chemical-exchange processes; a dia-
gram of the method is shown in Fig, 1.
The initial mixture, containing Ainit and Binit moles of the isotopes in the mixture, is
separated in the first extraction stage into two fractions, one of which contains Al and gi
moles of the isotopes and is enriched in the isotope A being extracted. The second, the de-
pleted fraction, contains A,, and B1 moles of these isotopes. After this, the fraction which
has been enriched once already is subjected to a second enrichment, for which it is again
separated into two fractions, which contains AZ and Bz moles (the enriched fraction) and A2
and Bz moles of the isotopes, respectively. After going through the required number of ex-
traction stages, we obtain in the last enriched fraction a small amount of an isotope mixture
An and B with such a large difference in isotopic composition from the initial mixture that
it can be reliably detected with the currently available methods of analysis. To obtain an
even greater difference in the isotopic composition of the final samples, we can carry out
i
Translated from Atomnaya Energiya, Vol. 54, No, 1, pp. 46-49, January, 1983. Original
article submitted June 16, 1981; revision submitted August 3, 1982.
0038-531X/83/5401- 0045$07.50 ? 1983 Plenum Publishing Corporation. 45
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
a stage-by-stage extraction or the isotope mixture in the aepletea branch as well. to ao
this, we subject the depleted fraction of the first stage to a multistage depletion process
by separating it at each stage into an enriched and a depleted fraction, with removal of the
enriched fraction at each stage. In this case we analyze. the final enriched and depleted
fractions. The presence of two branches in the experiment increases the accuracy of the an-
alytic determination, but it makes the experiment much more complicated. Therefore, we have
considered only the calculation of the enriching branch.
Despite the fact that in various studies devoted to the determination of the coefficient
of separation of isotopes by multistage extraction the investigators considered essentially
the same scheme for the process, the calculation formulas differ greatly, since they reflect
different experimental conditions, and different assumptions are made in deriving the final
formula. Thus, in [4] the calculation formula is obtained for the enriching branch for the
case in which the initial amount of the isotope being produced is small and the amounts re-
moved at all the stages are equal:
t -1 -~- Y
where a is the isotope separation coefficient; t, ratio of the number of moles in the depleted
fraction to the total number of moles before the separation process in the stage under consid-
eration; n, number of extraction stages; and xo, yn, respectively, the mole fraction of one
of the isotopes in the initial mixture and in the final enriched fraction.
The authors of [3] propose a more accurate but also more difficult method of graphical
analysis. The graphical construction in the coordinates y and x is carried out for given
values of a and n; as a result, we can find the final. mole ratio yn of the isotopes in the
mixture. If the calculated y does not agree with the experimental value, the calculation is
repeated with a new value of a.
Another proposed calculation formula is a relatively simple one, in which we assume as an
approximation that the products of the mole fractions of the isotopes in the equilibrium frac-
tions at all the stages are equal [5]. For the. enriching branch we have
?/n-yinit
e= ~ ,
0
where y = 1/2(yn{"yinit)~ E = a - 1 is the enrichment factor; and Oi, number of moles of the
mixture which have been carried into the depleted fraction divided by the number of moles of
the mixture which enter the stage under consideration. To obtain the corrected enrichment.
factor e', the value of e calculated by formula, (2) must be divided by 1 - ~, i.e.,
E, - E (_3
1-e'
In calculating the enriching branch, the product of the mole fractions of the isotopes
in the equilibrium fractions is also set equal to the product of the average values of the
mole fractions of the isotopes in the enriched fractions.
All the approximations made. in the calculation formulas lead to corresponding errors in
the calculations. In the proposed method for approximately calculating o we assume not that
the amount of mixture removed is constant, as in [4], but that the amount of the isotope he-
ing extracted is constant, although in the removal during the experiment and in the analytic
check of the removal we consider the sum of the isotopes, i.e., the assumption reduces to the
equality OAB ~ ~A?
The initial quantities for calculating a are the constant amount of isotope A removed
into the second phase (~A = Ai/Ai_1 = const) and the total enrichment in isotope A in the
first phase after n extraction stages (QE). To derive the calculation formula, we set up the
following system of equations describing the conditions under which the extraction process is
carried out;
AiBi
BtAt
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
f'l ~
B~ AZ_1
Ai
?A= A? ,
:-~
A:-1= As + Az
In this system Eqs. (4) and (5) determine a and ~i at stage i, and Eqs. (7) and (8) des-
cribe the balance with respect to each of the isotopes. Solving this system, we find
The total enrichment is determined by the following equations:
An Birth- Ai Binh Az Bi An Bn_1 R
Bn A init Bi A init Bz Ar Bn An-r
From Eq. (y) it follows that Qi at each stage is constant, since by the conditions of
the experiment, p A is taken to be constant. Consequently, QE = Si or
~: _ ~NE?
Taking account of (11), we can represent Eq. (y) in the form
If in the initial system of equations we replace Eq. (6) with the equation
nR- B` ~ (13)
B[_1
i.e., if we take the extracted amount of the second isotope to be constant, we obtain the
expression for a, in a different form:
?s ~~E
a = n
1- ~~E(1-?a)
In order to verify the above formulas and compare them with previously published methods
of calculation, we worked out a program of stepwise analytic calculation of the process of
multistage extraction which simulates the real process. Using the analytic method, we cal-
culated each extraction stage before the final isotope composition y was obtained. We then
compared with the Qiven value of yn all the yn obtained by differentncalculation methods. The
initial ~uantitiesfor the calculation are a, the separation factor in the chemical-exchange
system; Ainit and Binit~ the initial numbers of moles of the isotopes in the mixture; and
OAg, the relative amount of the isotope mixture extracted, at each contact in a given stage,
from the initial number of moles, into the other phase. In this case
At_1-{-B=_1
Taking account of (7) and (8), we can write Eq. (.4) in the form
Ai (Bi-r-Bi~
=a
"Bi'~A i-~ =Ai)
Making use of (7) and (8), we can easily transform Eq. (15):
A(-1'~"Bt-1 oA73r
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 t Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
?1
sub.'~i~u~.i iab iiVau Y. ~1V/, WC VUIQlLI
1 - _ Ai _ _ ``li Bi-i
Ai-i-f'Bi-1 (Ai~'a(Ai-i-A?))(Ai-i'~'Bi-i)
Equation (17) can be reduced to the form
kAi -~- bAt -I- e = O>
from which we find A..
i
Thus, our stepwise analytic calculation consisted of the following: Knowing the initial
amounts of the isotopes in the mixture, Ainit and Binit~ we determined Al and B1, then AZ and
BZ, etc.
To simplify the calculations, as in [5], we took (0~)i = (OAB)~ = 0.5 and e = 0.05 and
0.1 for a number of initial isotope concentrations. The calculations were carried out for
formulas (1), (2), (3), (12), and (14), and also for formula (14) using the total enrichment
(3'E instead of Q~: .
For each formula investigated, we calculated the enrichment factor e = a - 1, with the
aid of the previously obtained stepwise analytic calculations of the values yn, i.e., the final
isotope concentration in the mixture, and we also determined the relative error characterizing
the accuracy of the formula 8e = (e- eo/eo)?100%; we found eo from the conditions of formula
(17) of the stepwise calculation. The calculated results are shown in Table 1. It can be
seen that the accuracy of the calculation by formulas (2), (3), (12), (14), and (19) is af-
fected not by the regions of isotope concentration but mainly by the separation factor a (or
e). When a increases by a factor of two, the accuracy of the indicated formulas is reduced
also by a factor of almost two. On the other hand, the accuracy of formula (1) decreases
sharply as a decreases. Moreover, it can be said that for any of the isotope concentrations
considered, the highest accuracy is obtained by using formula (14) with the correction (19) -
the largest relative error is +1.4%, which is little more than half the relative error of the
other formulas.
Thus, it can be observed that the assumption OAB ~ 0,~ made in deriving formulas (12) and
(14) leads to smaller errors in the calculations than the assumptions in formulas (1)-(3).
We can estimate the error introduced by these assumptions, i.e., show the connection between
pA~and OA, by using Eq. (17). Dividing the numerator and the denominator of the second term
by Ai_1 and of the third by Ai_lAi_1j and writing Ai/Ai_1 = 1 - OA, and Bi_1/Ai_1 = K, we can
write Eq. (17) as follows:
1 - 1+K -{(q-OA)-}-a (1-(1-?A)]}(1+K) oAB?
After transformation, this equation reduces to the form
If, for example, sae take Bi_1 = 0.93, Ai_1 = 0.07, a = 1.02, and 0~ = 0.5, then the value of
OA is equal to 0.4937, i.e., when 0~ is replaced by 0 A, we have an error of 1.25%, which
does not exceed the error of the usual chemical analysis conducted to determine 0~ at each
stage.
Making use of Eqs. (12) and (14), we can take optimal conditions for conducting the ex-
periment with multistage extraction for each concrete case. The choice of the conditions is
determined by the accuracy of the isotope-analysis method, by the final amount of material
required, and, lastly by the practically most acceptable ratio between the amount extracted
at each stage and the number of stages.
For such an analysis we must specify the total enrichment sE. It must be sufficient
for obtaining the required accuracy. For example, we could specify QE = 1.05.
One of the main problems that must be solved in carrying out the multistage extraction
experiment is the choice of the optimal ratio between the amount of material extracted at
each stage and the necessary number of stages. This ratio is determined by the equation
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
Ainit+ Binh I I -
~ A,. B; 2 Az,Bzl i-7 A,.z,B~-~ i A,B1.9~.,;B?-; ~ A~,Bn
I I
I I
A,, B, Az. Bz I ~t i. B; f -. -Ai B< J A.~~ Ba
Fig. l.. Diagram of the process of multi-
stage extraction of an isotope mixture.
Num-
I OZ e.
Ulnii
7~n
ber of
10"- E
F_F.p ~
100 %
E
7 OZ e
F,-Ep
- 100%
E
102 E
E--Ep
- 100%
e
102 e
E_Ep
100"~
~ OZ E
F-E~ 1 OO?Aa
10"- E
E-E~ 100%
given
,
StdgeS
(1)
~(9)
(2)
o (2)
(3)
~ (3)
(12)
E0~,12~
(14)
Eo~i4~
(14)
e~~10)
5
0,0753
0,1048
15
5,21
-~-4,9
4,80
-3,996
5,04
~-G,84
4,89
-2,00
4,89
-2,0
5,02
-11
4
5
0,2619
0,39059
90
4,33
-93,4
4,84
-3,98.
5,09
-~1,7
4,91
-1,84
4,91
-1,8
5,113
,
-{-0
6
10
0,2624
n,32u71
G
9,72
-2,79
9,4
-5,9
10,39
?i-3,87
9,66
-3,4
9,66
-3,4
10,14
,
x-1,4
0 0, 7 0, H 0, 6 B
Fig. 2. Relation between the number of stages and
the amount extracted at each stage when (32 = 1.05
for: l) a = 1.01; 2) cti = 1.02.
which follows from (12). We can consider two systems with a = 1.01 and a = 1.02. Figure 2
shows the two curves in the coordinate, n - 0 corresponding to these systems. As can be seen,
for the system with a = 1.01 the total enrichment SE = 1.05 can be obtained if at each stage
we extract 10% of the isotope mixture entering it. In this case the number of stages is 49.
A similar enrichment can be obtained if at each stage we extract 20% of the isotope mixture
in the case of 24 extraction stages, or 70% of the isotope mixture in the case of 7 stages.
For a more effective system of chemical exchange with a = 1.02, the enrichment Q2 = 1.05
obviously can be attained by extracting 10% of the isotope mixture at each of 24 extraction
stages, or 70% of the isotope mixture at each of three stages.
These relations follow from the general relation between R and 0 in the case of one-time
contact. in this case, as can be seen from (22), for n = 1, the value of ~ varies linearly
from 1 to a when 0 varies correspondingly from 0 to 1.
It is obviously desirable to construct the multistage-extraction experiment in such a
way that the number of extraction stages will be a minimum. However, this is limited by the
neeessity of obtaining a specified absolute quantity of the material in order to conduct the
final isotope analysis with sufficient accuracy. For example, after 49 extraction stages,
with extraction of 10% of the isotope mixture at each stage, from an initial isotope mixture
of Ginit = 1000 moles there remains Gfin = 6.0 moles, and after 7 extraction stages with ex-
traction of 70%, there remains Gfin = 0.22 moles.
TABLE 1. Extraction of the Isotope Mixture along the Enriching Branch
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
Thus, by using the proposed equations (1L) and (14), we can conduct a preliminary analy-
sis of the investigated system of chemical exchange in order to choose suitable conditions for
the practical realization of the multistage extraction, and, determining S~ from the final
composition of the isotope mixture, we can find the isotope separation factor a
1. A. M. Rozen and A. I. Mikhailichenko, Zh. Fiz, Khim,, 64, No. 7, 1737 (_1970).
2, G. M. Panchenkov, V. D. Moiseev, and A. V, Makarov, ibid., 31, No. '8, 1851 (1957),
3. G. K. Boreskov and S. G. Katal'nikov, ibid., 35, No. 6, 1240 (1961),.
4. E. M. Kuznetsova, A. V. Makarov, and G. M. Panchenkov, ibid., 32, No. 11, 2641 (1958).
5. A. I. Mikhailichenko and 0. F. Shcheglov, ibid., 62, No. 1, 164 (1968).
V. N. Kadushkin, Z. P. Kiseleva,
G. A. Radyuk, B. G. Skorodumov,
I. I. Trinkin, V. A. Shpiner,
P. K. Khabibullaev, and V. N. Serebryakov
In recent years interest in investigation of the metal-hydrogen system has intensified,
this being associated with the solution of a number of problems in nuclear, thermonuclear,
and hydrogen power engineering. For research on the behavior of hydrogen in metals there has
been an intensive development of nondestructive nuclear-physical methods of measuring the
depth distribution of hydrogen in materials, using neutrons and especially ions as the analyz-
ing particles [1-4]. In our work we investigated the depth distribution of hydrogen in spec-
imens of copper, titanium, and palladium by the method of proton--proton scattering with re-
cording of the scattered particles in coincidences [2, 5].
The experimental setup for measuring the hydrogen concentration in the metals is shown
in Fig. 1. The specimens to be studied and standard targets are mounted on a rotary table
inside the scattering chamber. Figure 1 shows the case when the process of dissolution of hy-
drogen in palladium is investigated; a palladium wafer 100 um thick is vacuum-sealed in a
special cell which is filled with hydrogen to the required pressure.
A beam of 18-P1eV protons, collimated into a filament 1 mm in diameter, enters the cell
through an iron foil 30 um thick. Silicon-lithium detectors D1 and D2, with an effective
area diameter of 12 mm and a sensitive-region depth of 2 mm, and placed at an angle of ?45?
to the direction of the incident beam (angular aperture of detector ?0.1 rad), record events
of scattering on hydrogen contained in the palladium. Since coincidence events are selected,
scattering on the material of the specimen is present in the spectra only in the form of ran-
dom coincidences, since divergence of the scattered particles by an angle of 90? in the lab-
oratory system of coordinates is attained only with p-p scattering. A detector M of the same
kind was used to monitor the incident proton beam from the elastic peak of scattering from
iron foil 7 um thick.
The circuit for selecting coincidences consisted of two VF-3 time shapers, designed at
the Institute for Nuclear Research, Academy of Sciences of the Ukrainian SSR, a time-to-ampli-
tude converter, and a differential discriminator which controlled the linear transmission cir-
cuits in the spectrometric channels of the detectors. The energy spectra and the signal of
the monitor were fed into an IVK-2 measuring-computer, recording coincident events in a field
measuring 64 x 64 channels.
The time resolution of the pair of detectors was first checked with a thin hydrogen-con-
taining target and then was checked with a compound target which made it possible to give the
entire necessary energy range. The resolution was 5 and 18 nsec for the tin and compound
targets, respectively.
Translated"from Atomnaya ~nergiya, Vol. 54, No. 1, pp. 49-53, January, 19$3. Original
article submitted February 5, 1982.
50 0038-531X/83/5401- 0050$07.50 ? 1983 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
T..~ .,_~~~,.~?~~ LlllllL. 1L1C 1JV551U111Ly U1 (iCLCC:LlIl~' uyur.ugeu is ue~enniueu oy Lrie rela-
tion between real and random coincidences. It was shown in [5] that a detection limit of 10-6
H atoms/Me atom was attained by using an accelerator with a continuous particle beam. We
shall make an analogous estimate of this limit for the use of a cyclotron with a current-pulse
repetition rate v = 15 MHz, operating in a modulated regime with a duty factor p = 10. The,
load factor of each detector is proportional to the differential cross section 6p for the in-
teraction of the protons with the material of the specimen,
N = k6P, (1)
where k depends on the beam current, the solid angle of the detector, and the thickness of
the specimen,
The number of real coincidences is determined by the hydrogen scattering cross section
6H and the ratio of the number of hydrogen atoms to the number f of atoms of the matrix of
the specimen:
n=k6Hf. (2)
The probability of detection of random coincidences per second is
where F is the part of the detector load that falls- in the energy range in which cases of
scattering on hydrogen are observed (F ~ 0.1).
n f6H v _ f v QH
r Fzka~~ T~ FyN ~ aP '
For nuclei of average mass o H/o x 1 and with N 10? sec 1, we have
P
n
f ,.,10-4 nr , (5)
i.e., reliable determination of hydrogen (when real coincidences exceed random coincidences
tenfold.) is possible only at a level of 10-s H atoms/Me atom. The use of two-dimensional an-
alysis increases the sensitivity by approximately an order of magnitude because of the scat-
tering of random coincidences over the entire field. The advantage of using accelerators
with a continuous beam and the necessity of reducing the duty factor of pulsed accelerators
are obvious. With an increase in the detector load, the speed of the analysis improves, but
the ratio of real to random coincidences increases.
Let us point out, moreover, that under real conditions the number of coincidences is
always smaller than the value determined by formula (2). The number of real coincidences
is n only for an infinitely fine beam and in the case of absolutely exact alignment of the
beam-target~etectors system in ttie plane of the reaction. It is, therefore, necessary to
work with the narrowest possible beam and to align the system carefully.
Resolution and. Depth of Analysis. When: the detectors are set up at an angle of ? 45?,
the energy E of a detected proton is related to the depth of the location of the hydrogen
atom on which scattering occurs:
X- (2~+i/`' _ 1)-1 [t2~+il~ _a (Fo-2~E~)]> (6)
where a and Q are constants from the known relation between the range of the particle and
its energy for the given material (.R = aE~); Eo is the energy of the incident protons; t,
thickness of the specimen; and X, depth at which. proton collision occurred. Differentiating
Eq. (6), we obtain
whence it follows that the depth resolution (energy shift dE when the depth changes by dX)
depends on the energy E of the detected protons (or on the coordinate X). .Formulas (6) and
(7) enable the measured energy spectrum to be converted to the depth distribution by means
of the relation
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
Fig. 1. Experimental setup: 1) Faraday cylin-
der; 2) protons; 3) to supply system; 4) col-
limator.
N (X) = N (E) dE/dX (E). (8)
At an incident proton energy of 20 MeV for titanium the depth resolution with a spectro-
meter energy resolution of 200 keV will vary from 3 to 13 um. In a real situation, however,
the resolution is worse and varies little over the depth of the specimen. This is explained
by the straggling of the energy losses and the influence of multiple scattering with a finite
angular aperture DO of the detectors in the horizontal plane. Let us elucidate the latter.
Because of multiple scattering, a particle can deviate from the initial trajectory and
leave the specimen at a different angle within the limits of the angular aperture of the de-
tector. This leads to different energy losses of particles emerging in different directions.
If the mean free path of the particles in the specimen is Z, then the maximum difference of
the lengths of the trajectories is Z40 and, therefore, the energy of particlPG emerging from
one and the same point of the specimen can differ by DE = Z4~s, where S is the mean stopping
power of the material of the specimen. When Z = 200 Um and 4~ = 0.1 rad, the maximum change
in the energy losses because of this reaches 500 keV.
Meanwhile, it can be easily shown that the maximum path difference in the material of the'
specimen in the vertical plane is Z(o tt
is used for the readings of the neutron counter, where n. denotes the number of counts in the
j-th analyzer channel, and Zi denotes the analyzer channel number corresponding to the thresh-
old Ti (i = 1, 2, ..., m). The neutron spectrum ~(E) can be determined by solving the fo1-
lowing Fredholm integral equation of the first kind: '
Emax
N (T t)+6N (Tt)= I e (E, Tt) ~ (E) aE.
0
The notation is interpreted as follows: 8N (Ti) is the error of a measurement N(Ti); e(E, T.),
energy dependence of the neutron recording efficiency of the counter Sn with the threshold Ti;
and Emax, maximum neutron energy.
Equation (1) can be numerically solved on a computer with, say, the statistical regulariza-
ization method of [12]. In order to check the possibilities offered by the method, several
qualitatively different spectra were restored. The procedure is as follows. The selected
test spectrum ~t(E) was analytically given and the. values
Emax
A't (Tt)= ~ e (E> Tt) @t (E) dE
0
were calculated. Thereafter Eq. (1) was solved for ~(E) at
1V (Ti)=Nt (T t)?
The SN(Ti)/IV (Ti) values were assumed as 0.05, 0.10, and 0.15. Figure 1 shows the results
of a comparison for test spectra of the form
Translated from Atomnaya Energiya, Vol. 54, No. 1, pp. 68-69, January, 1983. Original
article submitted July 2, 1982.
80 0038-531X/83/5401- 0080$07.50 ? 1983 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
i i i I i i i i I i ~ i I~ I r ~ r r l~
100 200 J00 400 E, MeV
c,=~
c2= ~0
c,=~
cZ =2~0
~~i ~i ~~ i ~ I II
900 200 300 400. E, MeV
_Lt i i i ~ i ~ ~ i ~ i i i
300 400 E, MeV
Fig. 1. Results of restoring the neutron test spectra given by the function of~Eq.
(3) for SN(Ti)/N(Ti) = 0.15: a) counter readings not randomized; b) counter read-
ings randomized (g = 0.15).
~t (E)= l1-)-C, exp C- / E-Cz )ZJJ E-1
where C1 = 0 and 7, CZ = 70 and 210, and C3'= 0.8. In order to verify the stability of the
solution, the same spectra were restored with randomized Nt(Ti) and SNi(Ti) values which
were determined from a recurrence relation simulating the correlation between the various
N(Ti) values. The correlation of N(T.) results from the method of defining N(Ti) through
the instrument-dependent spectrum of ~he analyzer:
Nt (Tm) =Nt (Tm)-f-PmeNt (Tm)~
Nt (Ti)=Nt (Ti+i)-~-Nt (Ti)-
-Nt (Ti+t)~'Pig INt (Ti)-Nt (Ti+i));
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
SNt (T m)=b'n't (Tm)i
S1Vt (Ti)=V 8a Ni (Te+t)-)-8~ INt (Ti)-1Vt (Ti+t)l2~
i=m-1, m-2 ... 2, 1,
where p. denotes the values of a random quantity with a normal distribution, average zero,
and dispersion one; we have Ipil G l; and g = 0.05, 0.10; and 0.15. A comparison of the test
spectra with the restored spectra has shown that usually the calculated spectra correctly
describe the given test spectra within the error limits of the restoring method. Therefore,
there is reason to assume that the proposed method can be used for neutron spectrometry in
scattered radiation fields and at neutron energies of 15 to 500 MeV. The upper limit of the
energy range is given by the lack of .accuracy in calculating the efficiency of the neutron
detector at neutron energies above the threshold of meson generation and also by the form of
the response function at high energies. Estimates show that for neutron energies below 500
MeV, the contribution of ~ mesons, generated in the scintillator, to the neutron detector
efficiency does not exceed 5%.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 \ J
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
MEASUREMENT TECHNIQUES
Izmeritel'naya Tekhnika ,t
Vol. 25, 1982 (12 issues) ............................ $400
MECHANI~S OF COMPOSITE MATERIALS
Mekhanika~Kompozitnykh Materialov ~ ~
Vol. 18, 1982 (6 issues)? ....................; ......... $330
METAL SCIENCE AND'HEAT TREATMENT ,
Metallovedenie i Termicheskaya Obrabotka Metallov
Vol. 24, 1982 (12 issues) .......:.:... r.:......:..... $420
METALLURGIST
Metallurg ~ ; ~ ,
Vol. 26, 1982 (12 issues) ......................:......$435
PROBLEMS OF INFORMATION TRANSMISSION
- Problemy Peredachi Injormatsii ,
Vol. 18, 1982 (4 issues) ......... . .................... $320
~. .
PROGRAMMING AND COMPUTER SOFTWARE
Vol. 8, 1982 (6 issues) ' .......... $135
PROTECTION OF METALS" t ,
'Zashchita Metallov
Vol. 18, 1982 {6 issues) ..............:.........:....: $380
' ~ ,
RADIOPHYSICS AND QUANTUM ELECTRONICS
Izvestiya Vysshikh Uchebnykh Zavedenii, Radfofizika ,
Vo1.25,1982 (12 issues) .....:...................... $400
REFRACTORIES
Ogneupory ~ i
Vol. 23, 1982 (12 issues) .......~ ..................... $380
SIBERIAN MATHEMATICAL JOURNAL
Sibirskii Matematicheskii Zhurnal , ~ '
Vol. 23, 1982'(6 issues) ..........? .......:.......:.... $495
SOIL MECHANICS ANQ
FOUNDATION ENGINEERING ?
Osnovaniya? Fundamenty i Mekhanika Gruntov
Vol.-19, 1982 (6 issues) ........:..................... $380
t ?r
SOLAR S7~'STEM RESEARCH
Astronomicheskii Vestnik
Vol. 16, 1982 (4 issues) .:..........:.:.....:.....'..... $275
SOVIET APPLIEIy MECHANICS ?
Prikladnaya Mekhanika ,
Vol. 18, 1982 (12 issues) .......'..........' .:......:.. $400
Atomnaya Energiya
Volss: 52=53(12 issues) ......................~:.....:.. $440
~ ..
SOVIET JOURNAL OF GLASS'PHYSICS
AND CHEMISTRY ~ ?" ,
Fizika i Khimiya Stekla + t
.Vol. 8, 1982 (6 issues) ...... .......... $175
.............
t
SOVIET JOURNAL OF
NONDESTRUCTIVE TESTING . -
tDefektoskopiya
Vol. 18,1982 (12 issues) .............................$485
~ f ,
SOVIET MATERIALS SCIENCE
Fiziko-khimicheskaya Mekhanika Materialov
Vol. 18; 1982 (6 issues) .......' .............:......... $345
SOVIET MICROELECTRONICS
Mikroelektronika
r , ~ ~ ~
SOVIET MINING SCIENCE
Fiziko-tekhnicheskie Problemy Razrdbotki r
Poleznykh Iskopaemykh ;
.Vol. 18,1982 (6 issues) ...' ..........:.... ~ . $420
? ,.
SOVIET PHYSICS JOURNAL
Izvestiya Vysshikh Uchebnykh Zavedenii, Fizika .
SOVIET POWDER `METALLURGY AND
METAL CERAMICS
Poroshkovaya Metallurgiya
Vol. 21, 1982 (12 issues) ............................ $435,
STRENGTH OF MATERIALS ~ '
Problemy Prochnosti ~ ,
Vol. 14, 1982 (12 issues) ? .............:..:~..:......... $495
,, _
THEORETICAL AND MATHEMATICAL PHYSICS
Teoreticheskaya i Matematicheskaya Fizika
Vols. 50-53, 1982 (12 issues) .....~....! ............:. $380
,~
UKRAINIAN MATHEMATICAL JOURNAL
Ukrainskii Matematicheskii Zhizrnal ~, '
Vol. 34, 1982 (6 issues) .......:.~ ...................... $380
Send for Your Free Examination Copy
Plenum Publishing Corporation, 233 Spring St., NewrYork, N.Y. 10013
In United Kingdom: 88/90 Middlesex St:,~London E1 7EZ, England ,
Prices slightly higher outside the U.S.. Prices subject to change without notice.
Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
 Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9
-
RUSSIAN JOURNALS IN THE PHYSICAL
..
r .AND MATHEMATICAL SCIENCES
AVAILABLE IN E1~GLiSH 'TRANSLATION
ALGEBRA AND LOGIC ~ ~ l HYDRO1cECHNICAL CONSTRUCTION
Algebra i Logika ~ , ? Gidrotekhnicheskoe Stroitel stvo
Vol. 21, 1982 (6 issues) `. $270 Vol. 16 1982 (12 issues) ......................:....:4 $305
. ~ ~
ASTROPHYSICS INDUSTRIAti LABORATORY ~ ~, '
Astrofizika ~ ` ~Zavodskaya Laboratoriya . -
Vol. 18, 1982 (4 issues) . . ....:...:........... ~. >,...... $320 Vol. 48, 1982 (12 issues) .r'........................?.. x$400
AUTOMATION AND REMOTE CONTROL INSTRUMENTS AND ?
Avtomatika i Telemekhanika ~ 'EXPERIMENTAL TECHNIQUES .
Vol. 43, 1982 (24 issues) issues) ...'....` .............. $495., Pribory i Tekpnika ~ksperimenta
~ - Vol. 25, 1982 (12 issues) ....... ..................... $460
COMBUSTION, EXPLOSION,~AND SHOCK WAVES 's.
Fizika Goreniya i Vzryva _ ' ` JOURNAL~OF APPLIED MECHANICS '
Vol. "18 1982 6 issues ....;'$345 AND TECHNICAL PHYSICS ~
' ~ _ Zhurnal Prikladnoi Mekhaniki i Tekhnicheskoi,Fiziki
COSMIC RESEARCH Vol. 23, 1982 (6 issues) ............................>. $420
Kosmicheskie Issledov'aniya -
Vol. 20, 1982 (6 issues) ...................?.~....... , ... $425
CYBERNETICS -
Kibernetika - ~
Vol. 18, 1982 (b issues) ...:.........~:.............,..~.. $345
JOURNAL OF APPLIED SPECTROSCOPY
Zhurnal Prikladnoi Spektroskopii '
Vols: ~36-37 (12 issues) ..... . ............:............ $420
_ .
JOURNAL OF ENGINEERING PHYSICS
Inzhenerno fzicheskii Zhurnal ,
? .' ~ , _ Vols. 42-43, 1982 (12 issues) '...:' ..................... $420
DIFFERENTIAL EQUATIONS. , ~ .
Dijjerentsial'nye Uravneniya - ~ 'JOURNAL OF SOVIET LASER RESEARCH
Vol. 1'8, 1982 (12 issues) ?......:.. ........ $395 -
' ' ' ' ' ' ' ' ' ' ' A translation of articles based on the best Soviet research in the
DOKLADY BIOPHYSICS . ~ ~ ,
Doklady Akademii Nauk SSSR -
Vols. 262-267, 1982 (2 issues) ....................... $145
FLUID DYNAMICS'
Izvestiya Akademii Nauk SSSR, -
Mekhanika Zh'idkosti-i Gaza , -
Vol. 17, 1982 (6 issues) .~......, ...................... $380
FUNCTIONAL ANALYSIS AND
ITS APPLICATIONS ,
Funktsional:nyi Analiz i ~'go Prilozheniya
LITHUANIAN MATHEMATICAL JOURNAL
Vol. 16, 1982(4 issues) .........................:....... $320 ,
GLASS AND CERAMICS
Steklo i Keramika ~'
' Vol. 22, 1982 (4 issues) ....:.......... ' ............'.. $205
~ ` ~ 1. /
~ ?
Vol. 39, 1982 (6 issues) ..... ~ ... . ....::......... ~ ..... $460
HIGH TEMPERATURE ~ ,
Teplofuika Vysokikh Temperatur ~ __
Vol. 20, 1982 (6 issues}......... ...................... $400
field of lasers
Vol. 3, 1982 (d issues) ..:.:... ~'
................:....... $95
JOURNAL OF SOVIET 11~IA'THEMATICS
A translation of ~togi Nauki i Tekhniki and Zapiski
Nauchnykh Seminarov Leningradskogo Otdeleniya - ~ i
Matematicheskogo Instiluta im. ~V. A, Steklova AN SSSR_
Vols. 18-20, 1982 (l8 issues) .......................: $680
LITHOLOGY AND MINERAL RESOURCES
Litologiya i Poleznye Iskopaemye ~' - ~
Vol. 17; 1982,(6 issues) ......................:....... $420
MAGNETOHYDRODYNAMICS
Mdgnitnaya Gidrodinamika '
Vol. 18, 1982 (4 issues) .'... ~ ........................ $325
MATHEMATICAL NOTES , ~ ~ ~
Matematicheskie Zametki
Vols. 31-32, 1982 (12 issues) .......,.................. $400
' Declassified and Approved For Release 2013/02/06 :CIA-RDP10-021968000300020001-9