SOVIET ATOMIC ENERGY VOL. 46, NO. 2

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP10-02196R000800010002-4
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 86
• Document Creation Date: December 23, 2016
• Document Release Date: February 12, 2013
• Sequence Number: 2
• Publication Date: February 1, 1979
• Content Type: REPORT

Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
JJINJlitS-0,11 X
Russian Original Vol. 46, No. 2, February, 1979
SATEAZ 46(2) 81-160 (1979)
SOVIET
ATOMIC
ENERGY
ATOMHAH 3HEPIPIR
(ATOMNAYA gNERGIYA)
TRANSLATED FROM RUSSIAN
CONSULTANTS BUREAU, NEW YORK
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
SOVIET
ATOMIC
ENERGY
Soviet Atomic Energy is ,abstracted or in-
dexed in Applied Mechanics Reviews, Chem-
ical Abstracts, Engineering Index, INSPEC?
Physics Abstracts and Electrical and Elec-
tronics Abstracts, Current Contents, and
Nuclear Science Abstracts.
Soviet Atomic Energy is a cover-to-cover 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 betweenspublication 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.
- Editorial Board of Atomnaya Energiya:
Editor: 0. D. Kazachkovskii
Associate Editors: N. A. Vlasov and N. N; Ponomarev-Stepnoi
I. N.-Golovin
V. I. ll'ichey
V. E. Ivanov
V. F. Kalinin
P. L.. Kirilov,
Yu. I. Koryakin
- A. K. Krasin
E. V. Kulov
B. N. Laskorin
V. V. Matveev
I. D. Morokhov
A. A. Naumov
A. S. Nikiforov
A. S. Shtan'
B. A. Sidorenko
M. F. Troyanov
E. I. Vorob'ev
Copyright? 1979, Plenum Publishing Corporation. Soviet 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 personal or
internal use. However, this consent is given 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 U.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.
Consultants Bureau journals appear about six months after the publication of the
original Russian issue. For bibliographic accuracy, the English issue published 1:4
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 volume or pariicu-
lar issue of a Consultants Bureau journal, please. specify the-date and, where appli-
cable, the volume and issue numbers of the original Russian. The material you will
receive will be a translation of that Russian volume or issue.
Subscription (2 volumes per year)
Vols. 44 & 45: $130, per volume (6 Issues)
Vols. 46 & 47: $147.50 per volume (6 Issues)
Prices somewhat higher outside the United States.
Single Issue: $50 -
Single Article: $7:50
CONSULTANTS BUREAU, NEW YORK AND LONDON _
227 West 17th Street -
New York, New York 10011
Published monthly. Second-class postage paid at Jamaica, New York 11431.
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
SOVIET ATOMIC ENERGY
A translation of Atomnaya Energiya
August, 1979
Volume 46 Number 2
February, 1979
ARTICLES
Experimental Measurement of the Gravity Coefficient of Coolant Reactivity
for Reactors at the Bilibinsk Atomic Combined Electric Power and Heat-
Generating Plant -TA. V. Bondarenko, A. A. Vaimugin, P. G. Dushin,
A. G. Kostromin, G. V. Plotnikov, G. E. Soldatov, V. N. Sharapov.
CONTENTS
Engl./Russ.
and E. A. Yanovskii
81
75
Tritium Content in the Coolant of Water-Cooled?Water-Moderated Reactors
? L. I. Golubev, V. M. Ilyasov, A. I. Lur'e, B. N. Mekhedov,
L. N. Suk.hotin, V. M. Arkhipkin, and L. P. Kham'yanov
85
79
Nuclear Reactor Control by an Asymmetric Regulating System
? I. Ya. Emeltyanov, L. N. Podlazov, L. N. Aleksakov, and V. M. Panin . .
? ?
?
90
82
Theoretical?Experimental Model of the Nonsteady Radiational Creep
of Ceramic Fuel ? V. B. Malygin. Yu. V. Miloserdin, K. V. Naboichevko,
N. S. Golovnin, and Yu. K. Bibilashvili.
96
87
Effect of Irradiation Conditions and Chemical Composition on Radiational-
Damage Development in Steels and Alloys Irradiated by Neutrons
? V. I. Shcherbak, V. N. Bykov, V. D. Dmitriev. and S. L Porollo
101
91
Measurement of Neutron Spectra in Critical Assembly by Activation Method
? K. I. Zolotarev, V. P. Koroleva, Yu. F. Koleganov. and L. A. Chernov
106
96
Karmar-1 Pulsed Electron Accelerator with Relativistic Electron Beam Power of
up to 5 -1012 W/cm2? B. A. Demidov, M. V. Ivkin, V. A. Petrov,
and S. D. Fanchenko
111
100
Measurements of Dose Equivalent of Mixed Radiation outside the Serpukhov Proton
Synchrotron Shield? A. V. Antipov, I. S. Baishev. V. T. Golovachik,
G. I. Krupnyi, V. N. Kustarev, V. N. Lebedev, and M. Khefert
116
105
OBITUARY
Viktor Mikhailovich Gusev ? B. B. Kadomtsev, V. V. Orlov, M. S. loffe,
Yu. V. Martynenko, V. V. Titov, and 0. B. Firsov
121
109
LETTERS
Determination of the Absolute Yields of 43.5-, 74.7-, 117.8-keV y Photons
from 243AM ? Yu. S. Popov, D. I. Starozhukov, V. B. Mishenev,
P. A. Privalova, and A. I. Mishchenko
123
111
One Microwave Method of Dosimetry for Pulses of Penetrating Radiation
? V. N. Kapinos and Yu. A. Medvedev
124
112
Use of Thermal-Neutron Probes to Measure Thermal-Neutron Flux
of Distributions ? Yu. A. Satin, S. G. Karpechko, P. G. Afanas' ev,
V. I. Nalivaev, V. B. Pampura, and V. I. Uvarov
127
114
Two-Dimensional Modeling of the Fuel Assemblies of High-Temperature
Gas-Cooled Reactors ? M. D. Segal' and V. I. Khripunov
129
115
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
CONTENTS
Maximum Rate of Emission of Long-Lived y-Emitting Aerosols Allowable under Atomic
Power Station Standards and Control - G. G. Doroshenko, E. S. Leonov,
(continued)
Engl./Russ.
Z. E. Lyapiita, V. A. Fedorov, and K. N. Shlyagin
132
117
A Contactless Method of Studying the Thermal State of Fuel Elements during Irradiation
- V. N. Murashov, L. S. Kokorev, and V. V. Yakovlev. . . ? . . ...... .
134
118
Distribution of Tritium in Technological Systems of the Novovoronezh Nuclear
Power Plant - D. P. Broder, L. I. Golubev, V. M. Ilyasov, A. I. Luz.' e,
B. N. Mekhedov, I. R. Nurislamov, L. N. Sukhotin, L. P. Kahmiyanov,
and V. M. Arkhipkin
136
120
Effect of Helium-Ion Energy and Irradiation Temperature on the Blistering of Nickel
- V. I. Krotov and S. Ya. Lebedev
139
122
Kinetics of Instantaneous Neutrons in a System with a Cavity A. S. Chistozvonov,
I. P. Mat-veenko, V. P. Polivanskii, and G. M. Vladykov
140
123
Track-Detector Determination of Nuclear-Fuel Contamination of Primary-Circuit Sodium
Coolant - V. P. Koroleva, P. S. Otstavnov, and V. S. Shereshkov ?
143
125
BOOK REVIEWS
I. N. Aborina. Physical Research on Water-Moderated- Water-Cooled Power Reactors
- Reviewed by V. I. Pushkarev
145
126
JUBILEE
Yulii Borisovich Khariton (75th Birthday) - A. P. Aleksandrov, E. P. Velikhov,
E. I. Zababakhin, Ya. B. Zel' dovich, I. K. Kikoin, and M. A. Markov
147
129
INFORMATION
Work on Fast Reactors in Italy - V. M. ArIchipov
150
131
CONFERENCES, SYMPOSIA
Symposium on Hierarchy in Large Power Generation Systems - N. A. Trekhova
152
132
Conference on Large Tokamaks - G. N. Popkov
153
133
International Conference on the Application of the Mossbauer Effect
- A. N. Artem' ev
155
134
All-Union Problem Symposia on Real-Time Data-Computing Systems
- V. I. Vinogradov
156
135
NEW BOOKS
R. G. Bogoyavlenskii. Hydrodynamics and Heat Exchange in High-Temperature Pebble-
Bed Nuclear Reactors - Reviewed by V. P. Smetannikov
159
136
The Russian press date (podpisano k pechati) of this issue was 1/22/1979.
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 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
ARTICLES
EXPERIMENTAL MEASUREMENT OF THE GRAVITY COEFFICIENT OF
COOLANT REACTIVITY FOR REACTORS AT THE BILIBINSK ATOMIC
COMBINED ELECTRIC POWER AND HEAT-GENERATING PLANT
A. V. Bondarenko, A. A. Vaimugin,
P. G. Dushin, A. G. Kostromin,
G. V. Plotnikov, G. E. Soldatov,
V. N. Sharapov, and g. A. Yanovskii
UDC 621.039.524.2.034.44:621.039.519
The Bilibinsk atomic combined electric power and heat-generating plant (BATgTs) consists
of four identical units with channelized boiling water?graphite reactors whose heat is re-
moved by naturally circulating coolant through six independent loops locked to a drum-separa-
tor [1, 2]. The first unit was placed in operation in January 1974; the fourth was put into
service in January 1977.
All the reactors are of practically identical construction. The third and fourth reac-
tors differ slightly from the first two in that four (out of 60) control and safety rod units
are located in the active zone. The neutron-physical characteristics of the reactors are
therefore the same for the same fuel burnup.
The electrical and thermal output of BATgTs is regulated in accordance with a dispatching
plot. During a 24-h period the output varies by a factor of 7-8. The daily variation of the
load graph (ratio of minimum to maximum) reaches 0.59 on winter days and 0.68 on summer days.
In relation to this, it is important to know the parameters which determine the controlla-
bility of the reactors, in particular the reactivity coefficients.
For rapid transitional processes, such as during sudden changes of output, the reactivity
of the BATgTs reactors varies mainly because of the changes in coolant density and fuel tem-
perature. This is because the time constants of the changes in the other parameters which
affect the reactivity (temperature of the graphite walling, xenon concentration) are consider-
ably larger than the time constants of the variation in coolant density and fuel temperature.
Of all of the reactivity coefficients, the ones which are therefore of greatest interest from
the viewpoint of regulating these reactors are the gravity coefficient of the coolant and the
temperature coefficient of the fuel. In calculations, the greatest error is associated with
the gravity reactivity coefficient. It is therefore important to measure it experimentally
in order to improve the accuracy of the neutron-physical reactor characteristics.
From March 1976 to January 1977 a series of measurements was carried out at the BATgTs
in order to determine the gravity coefficient of the coolant reactivity and to clarify how
it varies during the course of an operating period in proportion to the fuel burnup. At that
time none of the BATgTs reactors had yet completely exhausted the reactivity established at
the initial loading, which was the same for all the reactors. There were, however, wide dif-
ferences among the reactors of the different units with respect to uranium burnup, since
, there was an interval of about a year which elapsed between the time each unit of the station
was placed into service and the time the following one was placed into service. The measure-
ments could be carried out in a rather short period thanks to the fact that several identical
reactors with different amounts of uranium burnup were in simultaneous operation at the sta-
tion. A total of seven measurements were made on the reactors of the first to the third units
in a range of values of average uranium burnup of 0.2-5.5 MW-day/kg U. It shouldbe noted that
at the end of the first operating period (,700 effective days) the average uranium burnup in
the production channels loaded in the reactor equalled 6.1 MW-day/kg U. The measurements
thus spanned practically the whole range of values of uranium burnup characteristic of the
first operating period of the reactors.
The measurements were made as follows. At a certain time there was a change in the
steam content and coolant density in the production channels of the reactors, which were
Translated from Atomnaya gnergiya, Vol. 46, No. 2, pp. 75-78, February, 1979. Original
article submitted February 2, 1978.
0038-531X/79/4602-0081$07.50 ? 1979 Plenum Publishing Corporation 81
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
E94
2 - ---
C'3- 86
254- 84
252- 82
250- 80
4248- 78-
246- 76
244-74
312 72-
?..3101-70,--%.?.
308 68 -
66 -
64 ?
62
-,640r 60-
a 6jQL 58-
-620 56-
c.510_54
608 52-
50
---
?????
0
/34
136
'38
140
142
144.
146
148
150
C7
730
720 =
1013:fission/cm9-sec and a = 2 kgf/mm2: 40 u02 in the Channel [1, 3, 10-13,
18]; 0) UO2 outside the channel (21]; A) (UPu)02 13, 14, 15]; -77-) (ITII)02 out-
side the channel [23];o) UN [18]; m) UC [12, 17].
Material in the volume of peaks of different ages is deformed at the same rate, but has
a different creep resistance. Considering parallel coupling Of the volumes containing peaks
of different ages, the time dependence of the radiational-creep rate may be obtained, taking
Eq. (3) into account, as follows
? (?vpain) , exp (-9:11:1) -n
71 o V pki,
V so )
11;Tt- 0
Equation (4) describes nonsteady radiational
occurs at the moment of reactor startup or later.
before reactor startup, the nonsteady radiational
(t) in the first term is replaced by E(T), where
tion of the load to the sample.
(4)
creep in the case when .the loading sample
In the case when the sample is stressed
creep may again be described by Eq. (4) if
T is the time elapsed since the applica-
As is known, the time dependence of creep deformation is well described by the Li rela-
tion [5]. Substituting this relation into Eq. (4), and setting
is obtained
yk (r LIDO)
e(D(t) rk+Lrel)? r (k ?1) exp ? (r up(D) ti
n = 1, the following result
(5)
where y is the steady rate of creep in the absence of irradiation; r is a coefficient charac-
terizing the duration of the nonsteady-creep stage; k is the ratio of the initial creep rate
to the steady rate.
For sufficiently large times, Eq. (5) transforms to the equation for steady radiational
creep
? 41,(r+,19
so? rk-11-vp0
Equation (5) is valid for'the whole temperature range, but
lowing relation is more convenient
ea, (t) =
vp:P+ v exp (?vpi:Dt) '
ccvv
(6)
for T < 900-1000?C the fol-
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
(7)
97
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
zph
????
?
100 ZOO 300. tO1
Fig. 3. Time-dependence of rate of nonsteady ra-
diational creep for UO2: instantaneous in-
crease in fission-Aensity,from zero to 10.22 fis-
sion/ce.sec; instantaneous decrease-in fis-.
sion density from 1022 fissionicm2-secto:.zero.-
where a = e r and v = kr are coefficients in the--logarithmic equation of creep-
0)==a1n(1-dhvO. (8)
Equation (7) is a particular case of Eq. (5)- for low temperature, when k >>0 And rt ?1.
For steady radiational creep at low temperatures when vp0-? v, the following-result is ob-
tained
? avono (9)
vpa+vCLUpCD.
?
As 0 -0- co, the radiational-creep rate does not exceed key at high temperatures and av at low
temperatures. At temperatures below 900-1000?C, the steady creep rate reaches saturation at
a fission density >1026-1027 fission/ceibsec, depending on the type of fuel [4].
The time dependence of the creep rate may be obtained from Eq. (5): for preliminary
stress of the sample
cvvjD
io(t)= vp04-vd-v exp (?opt) (up cDT ?1) '
where T is the time elapsed at the onset of irradiation; and for drop in reactor power
(10)
cwpcDv
co (t) (11)
vro?v [1 ? exp (?vp(1)-01+ tvvp(1) '
where t is the time after the drop in power; T is the total time of irradiation of the sample.
To evaluate the effect-of irradiation on the creep rate of a material, the parameter 8
was proposed in [4]:
tmax
8= 5 V p dt == vutE. (12)
where v(t) is the time dependence of the peak volume at a temperature T (0.5-0.7)Tp1; tilax
is the maximum time of existence of this peak; Vp.E. is the effective peak volume; tE is the
effective time of existence of the peak.
The temperature dependence of 8 given in [4] has now been corrected on the basis of
data on the thermophysical properties [6-9]. From Eqs. (12) and (9) there follows a linear
relation between the quantity 80 and the radiational-creep rate (Fig. 1). Data on the ra-
diational-creep rate of UO2, (UPu)02, UC, and UN were given in [111 for a stress of 1 kgff
mm2 and a density of 96% of the theoretical value. Thus, using Dia. (9) and (12), Fig. 1,
and the experimental value of a, the effective peak volume may be found for each fuel: vp.E.=
(1.40102)8.
98
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
fission/cm3
4,1f 3 4 5
11)-4 %
Kr6
10'
, --
200 400 600 800 1000 1200 1400 1800 1800 t,h
ot./01; fissic)nibni3
2 3
4
200 4170 MO. 800 WO t,h,
Fig 4, The dependence of nonsteady radiational
Creep with a = 2 kgf/mm2 and = fission/
cm' sec for UC (a) and taking account of increase
in fuel density for UO2 .(b): calculation;
---) experiment [12].
Investigation of Nonsteady Creep outside the Reactor
Radiation Field
If the relation given above is to be used for the calculation of the radiational creep,
information is required on nonsteady creep in the absence of irradiation. To this end, the
creep of 1/02 at a temperature of 293-1650K and a stress of 175 kgf/Mm2 was determined. To
predict the dependence of the radiational creep on the fuel density in the range'91-97% of the
theoretical density and on the grain size (from 1-3 to 15-20 1110., the effect of these factors
on the creep of UO2 was studied on a special apparatus [19]. The experimental data were
analyzed by the least-squares method by Eq. .(8) for low temperatures and by the relation
given in [5] for high temperatures (T > 1300?K).
The time dependence of UO2 creep deformation at 1073?C obtained in [20] for spiral coils
and recalculated for. -compressional deformation is satisfactorily approximated by Eq. (8). In-
vestigation shows that, within the limits of measurement error, the coefficient a is indepen-
dent both of the teat temperature in the range 293-4073?K and of the grain size, and increases
linearly with increase in stress up to 4 kgfimm2. The creep curve for 4 Sample annealed at
1673?K for 10 min is the same as the initial curve for the "fresh!' sample, and the value of
a is approximately the same in these experiments. Investigation shows a practically linear
:increase in the coefficient v, and hence in the rate of change of creep rate, with increase
in stress. The dependence of the creep rate on the density Of the samples has been associated
with the change in a in Eq: .(8), and approximated [11] by the relation
(z-.420+07.1249, (13)
where a0 is the coefficient at zero porosity; p is the relative porosity,
In investigating creep at high temperatures, the dependence of the Li-equation coeffi-
Cientsi5] on the stress and temperature was found. In deriving Eq: (3), it was assumed that
the creep mechanism does not change with change in the deformation,. since it has been shown
that the activation energy is independent of the deformation and the power index in the stress
is constant. More reliable prediction of the radiational-creep rate would require investiga-
tion of the effect of burnup on the coefficients of the equations describing creep in the ab-
sence Of irradiation.
99
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Radiational Creep of Ceramic Fuel
Analysis of the equations of radiational creep and measurement of the creep in the ab-
sence of irradiation provides the basis for a number of conclusions on the effect of various
factors on the radiational-creep rate.
The dependence of the radiational creep on the stress is analogous to the values ob-
tained for the rate of steady creep in the absence of irradiation at high temperatures and
for the parameter a in Eq. (8) at low temperatures. At a stress of less than 4 kgf/mm2, the
radiational creep of UO2 and (UPu)02 is directly proportional to the stress, which is in
agreement with experiment. The rate of radiational creep at low temperatures is independent
of the grain size and at high temperatures varies analogously to the rate of steady creep.
Knowing the effect of stoichiometry on the rate of steady creep in the absence of ir-
radiation, the rate of radiational creep at high temperature may be predicted. At low tem-
perature, the rate of UO2 radiational creep increases with deviations from stoichiometry,
according to preliminary data. At low temperature, the radiational-creep rate depends only
weakly on the temperature; with increase in temperature, the effect of irradiation on the
creep of ceramic fuel decreases (Fig. 2).
Nonsteady radiational creep is calculated from Eqs. (5), (7), and (11) using the para-
meters obtained for Eq. (8). The time dependence of the radiational-creep rate is shown in
Fig. 3 for the case when the calculated fission density is reached at the same time as the
sample is loaded or earlier (T = 0) and for the case of preliminary loading of the sample
(T = 50, 100, and 200); the parameters a and v are assumed to be constant. Preliminary load-
ing of the sample before constant fission density is reached may be identified with the pre-
liminary deformational hardening of the material in the course of tablet preparation. This
may be the explanation for the "nonstandard" time dependence of the creep rate observed in
[1, 13]. The time dependence of the radiational-creep rate for UC is shown in Fig. 4a. The
results of the calculations have been compared with the experimental data of [12]. The para-
meters a and v used may be constant at high burnup, when the change in structure and density
of the fuel has stabilized. Therefore, the curves in Figs. 3 and 4a are characteristic of
transient processes at high burnup. It follows from Eqs. (5), (7), and (11) that it is more
correct to take account of the variation in radiational-creep rate with fuel burnup.
Increase in density of the sample may affect radiational creep. On the one hand, in-
crease in density decreases a, which leads to decrease in the calculated radiational-creep
rate and increase in the time of the nonsteady process. On the other hand, increase in den-
sity results in decrease in size of the sample, and under compressional load this process
sums with the radiational creep (Fig. 4b). The increase in density was calculated from the
relationship obtained by analysis of the published data [11, 23, 25, 27]. The change in
sample size due to nonsteady creep calculated from Eq. (7), taking account of the change in
a with porosity, was summed with that arising as a result of increase in density. The re-
sults of the calculation are in satisfactory agreement with the experimental curve obtained
for UO2 [12]. This means that increase in density of the material under irradiation plays
an important role in the deformation of fuel.
LITERATURE CITED
1. A. Soloman, J. Am. Ceram. Soc., 56, Na. 3, 164 (1973).
2. D. Hough, J. Nucl. Mat., 52, 279 (1975).
3. D. Brucklacher, in: Proceedings of International CEBC Conference, Metals Society, Lon-
don -(1974), p. 118.
4. V. B. Malygin et al., At. Energ., 42, No. 1, 8 (1977).
5. J. Li, Acta Net., 11, 1269 (1963).
6. V. S. Chirkin, Thermophysical Properties of Atomic-Engineering Materials [in Russian
Atomizdat, Moscow (1968).
7. J. Leithaker and J. Godfrey, J. Nucl. Mat., 21, 175 (1967).
8. C. Affortit, J. Nucl. Mat., 34, 105 (1970).
9. H. Mikailoff, J. Cloude, and R. Lallement, in: Ceramic Nuclear Fuels, American Ceramic
Society, New York (1968), p. 113.
10. D. Brucklacher and W. Dienst, J. Nucl. Hat., 36, 244 (1970).
11. D. Brucklacher and W. Dienst, J. Nucl. Mat., 42, 285 (1972).
12. D. Hough, J. Nucl. Mat., 65, 24 (1977).
100
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
13. E. Sykes and P. Sawbridge, Irradiation Creep of Uranium Dioxide, Report RD/B/M-1489
(1969).
14. W. Dienst, J. Nucl. Mat., 61, 185 (1976).
15. D. Brucklacher and W. Dienst, in: Proceedings of IAEA Symposium on Experimental Results
on the Mechanical Interaction between Oxide Fuel and Cladding in Fuel and Fuel Elements
for Fast Reactors, Vienna (1974), p. 147.
16. D. Brucklacher and W. Dienst, in: proceedings of a Conference on the Physical Metal-
lurgy of Reactor Fuel Elements, Berkeley (1973).
17. J. Routbort et al., J. Nucl. Mat., 58, 78 (1975).
18. P. Zeisser, G. Maraniello, and C. Merlini, J. Nucl. Mat., 65, 48 (1977).
19. L.I.Laveikinet al., in: Radiation-Experiment Techniques [in Russian], No. 5, Atomiz-
dat, Moscow (1977), p. 24.
20. B. Burton and G. Reynolds, Acta Met., 21, 1073 (1973).
21. J. Perrin, J. Nucl. Mat., 39, 175 (1971).
22. J. Perrin, J. Nucl. Mat., 42, 101 (1972).
23. J. Routbort, N. Saved, J. Voglewede, J. Nucl. Mat., 44, 247 (1972).
24. N. Freshley et al., J. Nucl. Mat., 62, 138 (1976).
25. M. Marlowe, Trans. Am. Nucl. Soc., 18, 206 (1974).
EFFECT OF IRRADIATION CONDITIONS AND CHEMICAL COMPOSITION
ON RADIATIONAL-DAMAGE DEVELOPMENT IN STEELS AND ALLOYS
IRRADIATED BY NEUTRONS*
V. I. Shcherbak, V. N. Bykov, UDC 621.039.531
V. D. Dmitriev, and S. I. Porollo
The development of fast reactors has stimulated intense investigation in the field of
radiational materials science, as a result of which the main reatures of the development of
vacancy porosity and radiational creep have been established. At the same time, a very broad
range of questions remains unanswered, not least because the phenomena under investigation
are very sensitive to a large number of factors.
The present work attempts the analysis and generalization of the effect of neutron ir-
radiation on the properties of several experimental samples and also a number of steels and
alloys used in active-region components of the BR-10 and BOR-60 reactors. Preliminary re-
sults of these investigations have already been published [1-7].
Vacancy Pores
The investigation of the effect of a flux Kt on the swelling kinetics of OKh18N9T,
Khl8N9, 1Khl8N10T, and OKh16N15M3B steels has shown that in the initial period og vacancy-
porosity development the mean pore diameter increases according to the law (K01/2, regard-
less of the composition of the material (Fig. 1). The rate of pore growth then falls, and at
large fluxes the change in the mean pore diameter is slight. The pore concentration NV, in
contrast to the mean diameter, increases according to a power law up to approximately 100
dislocation/atom (TRN standard). Thus, for OKh18N9T steel, the relative pore volume at the
initial stage increases according to a near-quadratic law, while after 30 dislocation/atom,
AV/v increases in proportion to Kt.
Analysis of the experimental data shows that variable conditions of reactor operation
affect the kinetics of vacancy-pore initiation and growth [6]. The radiational swelling of
OKh16N15M3B steel used for fuel-element shells was less in the initial period of operation
of a BOR-60 reactor than in the case of irradiation in a BR-5 reactor. The features of the
irradiation of OKh16N15M3B steel in a BOR-60 reactor in this period include considerable os-
cillations of the temperature and rate of dislocation production. Nevertheless, in the
*This is an adaptation of a paper read to the Conference on Reactor Materials Science, Alush-
ta, 1978.
Translated from Atomnaya gnergiya, Vol. 46, No. 2, pp. 91-96, February, 1979.
0038-531X/79/4602-0101$07.50 0 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
101
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
1016
c)
,
014
600 ?
100
1 1 I
1 2 4 8810 20 40 60 100 ZOO
Kt, dislocation/atom
Fig. 1
0 2
6 810 20 40 60 80 100
1( 1, dislocation/atom
Fig. 2
Fig. 1. Effect of flux on NV and for pores in OKh18N9T (D), Kh18N9 (D),
OKh16N16M3B 00, and 1Kh18N1OT (A) steels irradiated at 460-470?C.
Fig. 2. Dependence of the total dislocation density in OKh18N9T (D), Kh18N9 01),
OKh16N15M3B (0), and 1Khl8N1OT (A) steels on the flux (Trad = 460-470?C).
complex changes in reactor conditions, it is possible to isolate characteristic values of
the power at which the reactor has been operating for a predominant part of the time. The
radiation-condition characteristics, results of investigation of this steel, and values of
the swelling calculated from the empirical expression of [7] are given in Table 1. It fol-
lows from Table 1 that for a certain time (at a flux of q,3-4 dislocation/atom) the material
was irradiated at a temperature less than 360-370?C, as a result of which the development of
vacancy porosity was considerably suppressed. In this case preliminary low-temperature ir-
radiation has the same effect on the swelling as preliminary cold deformation. With increase
in irradiation temperature, the annealing of the defects formed as a result of low-tempera-
ture irradiation becomes more rapid, and the effect of low-temperature is then less signifi-
cant.
Dislocation Structure
The investigation of the dislocation structure is of exceptional interest, since the
dislocation density determines the rate of initiation and growth not only of dislocation loops
and vacancy power but also of inclusions of other phases. The evolution of the dislocation
structure depends in a complex manner on the initial dislocation density Nd, the irradiation
conditions, and the chemical composition of the steels and alloys. In the initial stage of
irradiation, up to a flux of q,10 dislocation/atom, the total dislocation density increases
as a result of the development of dislocation loops, as follows from Fig. 2. Analysis of
the experimental results and comparison with data on the development of porosity shows that
the kinetics of dislocation-loop initiation and growth in steels irradiated by neutrons is
less sensitive to the change in chemical composition of the material and the initial disloca-
tion density. At relatively low temperature (",360?C), Frank interstitial-type dislocation
loops are observed in austenitic steels. With increase in irradiation temperature, prismatic
loops are also observed. In samples irradiated at >550?C, only prismatic loops are found.
Increase in nickel content in alloys of the system Fe--Cr--Ni leads to a sharp increase in
the number of prismatic loops [3].
As in the case of vacancy pores, the dependence of the mean dislocation-loop diameter
on the flux is characterized by two stages (Fig. 3). In the first stage, at small
fluxes, accelerated loop growth is observed. At fluxes above 5 dislocation/atom, there is
little change in the mean loop diameter.
The results of investigating the effect of the irradiation temperature on the kinetics
of dislocation-loop growth in OKh16N15M3B steel indicate that with increase in temperature
from 430 to 570?C the power-law index increases from 0.3 to 0.6 [6].
According to electron-microscope data, the increase in Nd at a flux of '1,10 dislocation/
atom is associated with the appearance in the structure of the steel at this time of
102
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
TABLE 1. Irradiation Conditions and the Swelling of OKh16N15M3B Steel in a BOR-
60 Reactor
Kt,
dislocation/
atom
T? c
Kt,
dislocation/
atom
T? c
Kt,
dislocation..!
atom
7.? c
Total flux, dis-
location/atom
AV/17;? %
3,1
334
6,3
395
31,6
425
41
0,05; 0,4
3,1
340
6,3
415
31,6
450
41
0,2; 1,0
2,9
360
5,9
450
29,6
490
38,5
1,0; 2,0
2,6
365
5,25
470
26.4
515
34
1,4; 1,9
*The first value is experimental and the second calculated.
high-density barriers in the form of loops and inclusions, which leads to a considerable dis-
tortion of the dislocation lines. It is for this reason that the dependence of the disloca-
tion density on the flux for 1Kh18N1OT is sharper than that observed for OKh18N9T steel.
Further increase in the flux is accompanied by slowing of the growth of Nd and at large
fluxes the difference in the values of Nd for the different steels becomes insignificant.
The irradiation of material with an elevated initial dislocation density is a more
complicated situation. In Fig. 4, the temperature dependence of the dislocation den-
sity is shown for OKh16N15M3B steel in which the value of Nd before irradiation was
4.10" cm-2, which corresponds to 5-7% deformation. As is evident from Fig. 4, ir-
radiation of steel at a temperature above 500?C led to a recovery process, as a result of
which a drop in dislocation density was observed. On passing to large fluxes, the recovery
processes was impeded by the development of an ensemble of inclusions and dislocation loops,
and an increase in Nd was observed over the whole temperature range. Increasing the initial
dislocation density to 3.1011 cm-2 leads to the development of a recrystallization process,
which becomes pronounced at temperatures above 500?C [4]. Further increase in Nd facilitates
the acceleration of recrystallization.
Inclusions
Particles of inclusions of other phases are observed in the irradiation of steels even
at a flux of q,5.1020 neutron/cm2. Investigation of the dependence of the concentration and
mean volume of a titanium carbide inclusion in 1Kh18N1OT steel and of a niobium carbide in-
clusion in OKh16N15M3B steel on the flux has shown that the initial stage of irradiation is
characterized by increase in the concentration and volume of the developing inclusions (Fig.
5). It follows from Fig. 5 that the formation and growth of niobium carbide in OKh16N15M3B
steel is rather more rapid than the corresponding processes for titanium carbide in IKhl8N9T
steel. With increase in the flux, the concentration of the inclusions increases to a maxi-
mum and then falls according to the law Np (Kt)-1. It is found that the coalescence of
inclusions of titanium and niobium carbide and Laves phase has a different dependence on the
flux. With increase in irradiation temperature from 430 to 570?C, the power-law index in
the formula for niobium carbide varied from -0.4 to -0.95. The power-law index in the equa-
tion for the mean volume of these inclusions in the given temperature range increased from
0.6 to 1.2. The coalescence of Laves phase in this temperature range does not depend greatly
on the temperature and the flux. As is known, the coalescence of the inclusions in steels
may occur in two different ways. In particular, the inclusions forming as a result of ir-
radiation may vanish as a result of the diffusion of individual impurity atoms from small to
large inclusions through the matrix. According to the theory of diffusional coalescence
[8],, the kinetics of growth of particles of radius Rp is determined by the equations Rp
t1/2, Np rt, lit. If diffusion occurs over the dislocation lines, coalescence occurs more
slowly than for volume diffusion, and the mean si9 and concentration of the inclusions must
vary over time according the relations [9] Rp 0/7 and Np t-3/7
Turning to the experimental data, it is evident that the power-law index in the equa-
tions for the mean volume and concentraiton of niobium-carbide inclusions in OKh16N15M3B
steel irradiated at 430?C is 3/7 and ?3/7, respectively, as follows from the theory of coales-
cence with the diffusion of impurity atoms along dislocations. At higher irradiation tempera-
ture, the dependence of the concentration and mean volume of niobium-carbide particles on the
flux takes a form similar to that predicted by the theory of diffusional coalescence.
Accordingly, it may be suggested that at low irradiation temperature the most probable
mechanism of inclusion growth is particle coalescence by diffusion of impurity atoms along
dislocation lines. In the case of irradiation at higher temperature and low dislocation
103
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
1014
800
"c 400
^^
\' 200
100
I III
I I I
2 4 6 8 10 ?20 40 60 100
dislocation/atom
Fig. 3
to"
0
2
m mo
400 450 500 550
Temperature of irradiation. ?C
Fig. 4
Fig. 3. Dependence of Ni and for dislocation loops in OKh16N15M3B (D)
and 1Kh18N1OT (A) steels on the flux '(Trad = 460-470?C).
Fig. 4. Temperature dependence of dislocation density in OKh16N15M3B at a
flux of 5 (0) and 20 (0) dislocation/atom.
600
TABLE 2. Values of the Parameter (n ?
1) for Some Steels and Alloys
Material
(11-1)?1O-2
Material
(11-1)?1O-2
1Kh12M2S2
0,3
oKh16N15M3B
1,1
OKh18N9T
0,8
OKh16N403
0,15
1Kh18N1OT
1,0
OKh16N8OB
0,9
00Kh16N15M3B
0,75
density, coalescence of inclusion particles as a result of volume diffusion is more probable.
At intermediate irradiation temperatures and high dislocation density, both mechanisms may act
simultaneously.
Effect of Chemical Composition
Electron-microscopic observation of radiational damage in OKh18N9T, 1Kh18N10T, 00Kh16N-
16M3B, and OKh16N15M3B steels has shown that the effect of inclusions on pore initiation and
growth depends on the irradiation temperature. In the investigation of samples irradiated
at temperatures above 500?C, it is observed that some of the largest pores are associated with
inclusion particles, while the relative proportion of such pores increases with increase in
irradiation temperature. In addition, some of the larger incoherent inclusions, like grain
boundaries, have regions that are free from pores. According to the results of [2], at tem-
peratures above 500?C the inclusion particles affect the development of vacancy porosity in-
directly, by increasing the dislocation density. This increase may be explained by the pres-
ence in the slip planes of scattered microdisperse inclusions, which lead, in conditions of
supersaturation of the material with point defects, to considerable distortion of the disloca-
tion lines.
The results of measurements of the swelling, dislocation density, and particle density
of inclusions in steels and alloys irradiated at 460-500? by a flux of up to 14 inclusion/
atom [2, 3] are shown in Fig. 6. According to these data, the development of pores and dis-
location structure in OKh18N9T (1), 1Khl8N1OT (5), 00Kh16N15M3B (4), and OKh16N15M3B (6) steel
and OKh16N8OB alloy (3) is determined to a considerable extent by the formation of inclusion
particles. On the other hand, as noted above, the kinetics of inclusion-particle initiation
and growth depend on the development of the dislocation structure. However, for 1Kh12M2S2
ferrite steel (7) and OKh16N4OB alloy (2) no such correlation is observed. These results in-
dicate that changes in the chemical composition of steels and alloys have a complex effect
on the initiation and growth of point-defect aggregations. The effect of alloying elements
on pore initiation and growth may be explained by a number of factors: an increase in
104
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
100
80
50
1 2 4 6 8 10 20 40 50 100
Kt, dislocation /atom
Fig. 5
0,01
1014
o2
ii *
t.car3
Fig. 6
N6cni2
lo"
100
LI /0 9
1016
Fig. 5. . Concentration and mean volume of inclusions of titanium carbide (A),
niobium carbide (D), and Laves phase' (0) as a function of the flux (Trad =
470?C; K = 2.2?10-7- dislocation/atom.sec).
Fig. 6. The effect of the inclusion-particle concentration on the total dis-
location density (0) And swelling (0) of some steels and alloys at Trad =
460-500?C and Kt = 14 dislocation/atom (the steels and alloys corresponding
to the numbers on the curves are identified in the text).
point-defect recombination rate [10]; anomalous changes in the partial diffusion coefficients
of substituent atoms [11]; changes in the vacancy and interstitial-atom diffusion coeffi-
cients, in the defect packing energy, and in the surface energy; or changes in the concen-
tration of inclusions able to act as neutral sinks [12] or to alter the dislocation density
[2].
Analysis of results on the swelling of steels and alloys shows that none of these mech-
anisms can itself account for the whole set of experimental data. A more profound under-
standing of the role of certain alloying additives may be obtained if it is assumed that
changes in the chemical composition significantly alter the capacity of the dislocations to
capture point defects.
Table 2 gives values of the parameter (n ? 1) obtained in [2, 3]; n characterizes the
ratio of the surface corresponding to capture by dislocations of interstitial atoms and va-
cancies. As follows from Table 2, the parameter (n ? 1) is approximately equal to 10-2 for
most steels and is reduced to 10-3 only for ferrite steel and OKh16B4OB alloy, which are
characterized by very small swelling. Hence, the small swelling of these materials may be
explained by the small difference in the surfaces corresponding to point-defect capture by
dislocations.
Note that the effect of alloying elements on the swelling of a material is more clearly
expressed at small fluxes and low irradiation temperature. One possible explanation for this
is that irradiation of the material at large fluxes or high temperature leads to profound
changes in the composition of the matrix as a result of the formation of inclusion particles.
Many alloying elements, for example, Ti, No, Nb, C, and N, enter solid solution, and thus
their role in pore initiation and growth becomes less significant.
CONCLUSIONS
The investigation of the initiation and growth kinetics of pores, dislocation structure,
and particles of phase inclusions at fluxes of 1-100 dislocation/atom leads to the following
conclusions.
1. The dependence of the pore concentration in steels on the flux is described by a
power law; the power-law index depends strongly on the composition of the steel and the ir-
radiation temperature. The mean pore diameter in the initial period of irradiation varies
according to the law (K02/2; at large fluxes, it remains almost constant.
105
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
2. In all the steels investigated, at fluxes above 5 dislocation/atom, the total dis-
location density increases according to a power law; the power-law index varies from 0.4 to
1 depending on the composition of the steel and the irradiation temperature.
3. In the initial period of irradiation, a stage of accelerated inclusion-particle
initiation is observed, while at high fluxes they are seen to coalesce. The correlation of
the kinetics of inclusion-particle initiation and growth with the dislocation density and the
irradiation temperature has been established.
4. The effect of the chemical composition of the steel, which is particularly clearly
expressed at low temperature and small fluxes, is evidently due to the action of the alloying
elements either on the ability of the dislocations to capture interstitial atoms or on the
mechanism of the formation of new-phase inclusions.
LITERATURE CITED
1. V. I. Shcherbak et al., in: Problems of Atomic Science and Engineering. Fuel and Con-
structional Materials Series [in Russian], No. 1(6), lad. VNIINM, Moscow (1977), p. 14.
2. V. I. Shcherbak et al., in: Problems of Atomic Science and Engineering. Radiation-
Damage Physics and Radiational Materials Science Series [in Russian], No. 1(4), Kharkov
Physicotechnical Institute (1977), p. 83.
3. V. I. Shcherbak et al., in: Proceedings of an International Conference on Radiational
Effects in Breeder Reactor Structural Materials, Met. Soc. AIME (1978), p. 773.
4. A. N. Vorob'ev et al., J. Nucl. Energ. Soc., 14, No. 2, 149 (1977).
5. S. I. Porollo et al., At. Energ., 43, No. 3, 207 (1977).
6. V. I. Shcherbak, V. N. Bykov, and V. D. Dmitriev, Fiz. Met. Metalloved., 43, No. 2, 419
(1977).
7. V. I. Shcherbak et al., J. Brit. Nucl. Energ. Soc., 14, No. 2, 145 (1975).
8. I. M. Lifshchits and V. V. Slezov, Zh. Eksp. Teor. Fiz., 35, 47 (1958).
9. A. Ardell, Acta Met., 20, 61 (1972).
10. F. Smidt and J. Sprague, Scripta Met., 7, No. 5, 495 (1973).
11. H. Venker and K. Ehrlich, J. Nucl. Mater., 60, 347 (1976).
12. R. Bullough and R. Perrin, in: Proceedings of a European Conference on Voids Formed by
Irradiation of Reactor Materials, BNES, Reading (1971), p. 78.
MEASUREMENT OF NEUTRON SPECTRA IN CRITICAL ASSEMBLY
BY ACTIVATION METHOD
K. I. Zolotarev, V. P. Koroleva, UDC 621.039.51
Yu. F. Koleganov, and L. A. Chernov
Measurement of the energy spectra of neutrons is of great interest for both reactor con-
struction and refinement of computational methods and choice of constants. Quite refined
spectrum-measuring methods (time-of-flight, recoil-proton, etc.) could not displace an earlier
method, the activation method [1] because of its considerable advantages. The main advantage
for spectrometry is that it has small perturbations at the point of measurement and that it
can be used in fields of mixed radiation and can encompass practically the entire energy
range of the spectrum produced in the reactor. A serious drawback of the activation method
is the lack of sufficient accuracy in the spectrum determined. The success of the applica-
tion of the activation method to neutron spectrometry depends on the accuracy of determina-
tion of the reaction rate, the number of activated tracers with independent variation of the
reaction cross section, the accuracy of the reaction cross sections used, and the method of
reproducing the spectrum. In order to increase the accuracy of spectrum reproduction, there-
fore, it is desirable to refine both the experimental methods and the methods of reproducing
the spectrum.
Translated from Atomnaya gnergiya, Vol. 46, No. 2, pp. 96-100, February, 1979. Original
article submitted February 20, 1978.
106 0038-531X/79/4602-0106$07.50 ? 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12: CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
TABLE 1. Experimental Values of Reac-
tion Rates
Reaction
Reaction rate,
1Oz4 events/sec ?
nucleus
Cross-section
data used
23Na (n, 17)24Na *
2,938402+4,61'
151
24mg (n, p) 24Na
1 , 196 ? 106 -1- 4 , 6
[7]
22AI {n, p) "Mg
2,891.106+4,7
[6]
37A1 (n, a) 24Na
5,810-105+4,6
[61
56Fe (n, p) 56M-n
7,931.105+4,6
[6]
"Ni (n, p) Co55
7,850.102+4,7
[6]
(n, n,) iismin
Iii-sIn
1,179.108+5,5
[6]
233U (n, f) *
5,465-1010+9,6
[5]
235U (n, f) *
2,483-1010+8,8
[5]
238U (n, f)
2,767.106+8,9
[6]
23013u (n, f) *
2,716.1010+8,8
[5]
*Reaction-rate values given from cadmium
limit.
tError, %.
The objective of the present paper is that of using the activation method to measure
the neutron spectrum in a methodological critical assembly development of a GIN program for
unfolding the spectrum from activation data on the basis of methods presented in [2, 3], the
application of the inductive approach to the process of spectrum reproduction, and compari-
son of the results with the spectrum measured by direct spectroscopic methods.
Experiment. The neutron spectrum was measured on a methodological critical assembly
with a zirconium hydride moderator (the ratio of nuclear concentrations of hydrogen to 233U
is q,25). In the center of the core we determined the absolute value of the reaction rates ,
for 23Na(n, y) 24Na, 24Ng(a, p)24Na, ,,Al(n, p)
27Mg, "Al(n, a)24Na, "Fe(n, p)38Mn, "Ni(n,
-
222u(a, f), 229pu(n,
p)"Co, 1"In(n, n')1"In, f) by the activation method and for 2331.J(n,
f) and 238U(n, f) by an indirect method. The activation specimens had a diameter of 9-25
mm. They were irradiated in cadmium filters 0.5 mm thick. The induced activity was measured
with a Ge(Li) semiconductor spectrometer (sensitive volume 5 cm3, energy resolution 2.7 keV
for "Co) by the photoabsorption of quanta of a particular characteristic energy for each
reaction. After amplification, the pulses from the detector were analyzed with a 512-chan-
nel analog-to-digital converter (ADC) and stored in the magnetic immediate-access memory
(MIAM) of the Reactor Measuring Center (RMC). The data accumulated in the MIAM were printed
out by a BZ-15 digital printer or teletype. The measured amplitude y-ray spectra were pro-
cessed on a Nairi-1 computer. The procedure employed to calculate the area of the photopeak
was described in [4].
Because of the lack of suitable specimens the 2331.J(n, f) and 238U(n, f) reaction rates
were determined by an indirect method from the ratios 433/4" and a 33/a133 of the spec-
trum-averaged fission cross sections of the respective isotopes and tie cadmium ratios q,713
and R5 for233U and 233U, measured by KNT-3, KNT-5, and KNT-8 ionization fission chambers.
Cd
107
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
10-6 10 io-4 10 10-2 10 100
n, MeV
Fig. 1. Neutron energy, spectrum unfolded by using
the inductive approach:: e) unfolded spectrum; 0)
experimental spectrum obtained by direct spectro-
metric methods of spectrometry.
The measured absolute values of the reaction rates-are given in Table 1 along with the
total error (in the experiment and in the constants used) for a confidence coefficient of
0.95-0.90.
The Spectrum-Unfolding Procedure and the GIN Program. . The procedure employed in the
present paper to unfold the neutron.spectrum-was based on the SPECTRA mathematical method first
developed by Greer et al. [2] and improved by Fisher and Turi [3]. In the method described
the sought differential neutron spectrum50(E) is presented in the form of a continuous piece-
wise-linear function EO (E) and the energy dependence of the microscopic cross sections for
nuclear reactions is presented in the form of a continuous piecewise-linear function a(E)/E.
In this case the system of linear Fredholm integral equations of the first kind can be trans-
formed into a matrix equation
a--7-Q(1), (1)
where a is a column-vector of n elements which are measured values of the reaction rates Ai,
0 is a column-vector of m elements which are the values of the neutron flux density sought
at given energy points, and"Q is an (n x ni) matrix whose elements are definite integrals of
the reaction cross sections.
If the rows of matrix Q are separated into the corresponding elements of vector a and
if the resulting matrix is denoted by C, then Eq. (1) can be rewritten as
? (2)
where (1m) is a column-vector of n elements, each of which is equal to unity.
Since the spectrum is usually determined at a larger number m of energy points than the
number n of available activation data, the solution of Eq. (2) is mathematically indeter-
minate. Additional information is therefore necessary in order to obtain physically sub-
stantiated results. In this case, an initial spectrum490(E) is prescribed and among the
spectra which satisfy Eq. (2) we find a spectrumco (E) for which the fundamentals
.Emax (E)
[ (E) dE and l (Avi ? Ai)2
(E) _
Ernin
(3)
simultaneously assume a minimum.value. Mere, Emax and Emir, are the upper and lower limits
of the energy range in which the spectrum is determined, and ABi and Ai are the calculated
and measured values reaction rates in the i-th isotope in the spectrum sought.
To derive an iteration formula with allowance for the condition (3) we set up the error
function AI which characterizes the deviation between the calculated and measured reaction
rates in the spectrum sought, on the one hand, and between the sought and initial spectra on
the other:
41 = ECcDI ? (1n)rr F2 [C(Di ? ( I )1 + (01 ? (Do)T G2 (00 ?to), (4)
where G2 and F2 are the diagonal normalization matrices (det G 0, det F 0). Minimization
of the function AI with respect to 01 leads to
(Di = G-413 [KTF (1n) + GOol,
108
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
ol -W7 1076 vl , Iv- 10-1 To Err MeV
Fig 2. Neutron energy Spectrum, unfolded with initial 1/E
spectrum prescribedat 50 energy points at once A) initial
i/B spectrum; 4* unfolded spectrum; experimental spectrum
'obtained by direct spectrometric methods.
where B = 0J1(4., K and I is a unit matrix of order m.
A new
4)2 with t2.
spectrum.
will be of
function A.2 analogous to Eq-. (4) can be constructed by replacing to with 4)2 and
Minimization is carried out with respect to the next approximation t2 of the
If this.process is repeated, the expression for the sought spectrum after k steps
the form
---G-i./3[KTF(1?)+GIA,,I.
Equation (5) is the general formula of the iterative process which is continued until the
rms deviation between the calculated and measured values of the reaction rates reaches a
given value. The iterative process is also halted when the deviations between the calcula-
tion and measured values of the reaction rates reaches a given value. The iterative process
is also halted when the deviations between the calculated and measured reaction rates become
smaller than the relevant errors in the experimental data on the reaction rates.
The algorithms described here, including the method of controlling the rate of conver-
gence of the iterative process [3], was executed in the GIN program. The GIN program is
written in FORTRAN IV which is suitable for the translator of the ES-1030 computer and makes
it possible to reproduce the values of the neutron flux density at 50 energy points. The
maximum number of reactions used cannot exceed 30. The data on the reaction cross
sections are fed in from magnetic tape whereas the other data are fed in from
punched cards. The length of the objective module of the GIN program is %32K.
Spectrum Unfolding and Results. The inductive approach to the solution of the problem
is proposed with a view to enhancing the reliability of the unfolding of the energy spectrum
and reducing the influence of the choice of the initial information on the results of the
algorithm. As already mentioned, the linear Fredholm integral equations of the first kind
are approximated by a system of algebraic equations in matrix form (1). The proposed induc-
tive method makes use of the fact that the solution of Eq. (1) can be unique only if the num-
ber of points of the spectrum sought coincides with the number of activation data available,
i.e., if m = n. The principle of the inductive approach lies in the stage-by-stage unfolding
of the spectrum. In the initial stage the spectrum is determined at points which correspond
in number to number of experimental data. Each subsequent stage in the reproduction of the
spectrum is accompanied by a gradual increase in the number of subdivisions of the energy
scale. In any case, the results of the preceding calculation are used as the initial spec-
trum. In the problem of the initial spectrum at intermediate points account was taken of
the assumption of the piecewise-linear dependence of the function E (E). The process of
spectrum unfolding is continued until the resulting spectrum begins to repeat the spectrum
of the preceding stage or until the number of points on the energy scale does not reach the
maximum value predicted by the program.
It is known that the initial a priori spectrum is most expediently prescribed from
mathematical calculation of the medium studied. However, in order to verify the capabilities
(5)
109
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
of the inductive method in unfolding the neutron spectrum of U--ZrH of a critical assembly,
we deliberately used a 1/E spectrum differing substantially from the real spectrum, as the
initial spectrum. The spectrum unfolding was carried out in five stages. In the first
stage the 1/E spectrum was prescribed at 11 points on the energy scale (in accordance with
the available activation data) in the range from 0.01 eV to 18 MeV with approximately the
same step on the logarithmic scale, apart from the interval 8-800 keV. When the number of
subdivisions of the energy scale was increased further the results of the fourth and fifth
stages already practically did not differ from each other. Therefore, the spectrum after
the fifth stage of unfolding, which is shown in Fig. 1, should be considered final. For
comparison, we show the spectrum measured by direct methods [8] (using the time-of-flight
and recoil-proton methods). Both spectra were normalized with respect to the absolute values
of the reaction rates. The values of the normalization factor X was found from the condition
that the expression E(X4i,) have a maximum. It is seen from Fig. 1 that in the energy
ranges 0.5 eV-8 keV and 800 keV-18 MeV the spectra are in satisfactory agreement
with each other (the maximum difference is no more than 15%). Because of the inadequacy of
the experimental information no points were prescribed during the unfolding of the spectrum
in the energy range 8-800 keV. The curve was drawn with the assumption of the linear depen-
dence of the function Ecio (E) inside this range; it is meaningless, therefore, to speak of the
accuracy of the reproduction of the spectrum.
To get a graphic picture of the advantages of the inductive method over the traditional
approach to the reproduction of a spectrum (when the subdivisions of the energy scale far
outnumber the experimental data) we made a calculation of the spectrum with the initial spec-
trum prescribed at 50 points at once. The results of the spectrum unfolding are shown in
Fig. 2, from which it follows that the spectrum obtained does not give satisfactory results
if the initial spectrum differs substantially from the real spectrum. Thus, in the range of
intermediate energies the unfolded spectrum displays considerable oscillations whereas the
time-of-flight measurements yield a smooth curve which obeys the E-??86 law. The most pro-
nounced difference between the unfolded spectrum and the real spectrum is observed in the
energy range 6-400 keV in which the resulting spectrum essentially repeats the initial spec-
trum, this being explained by the lack of sufficient experimental data in this range. Com-
parison of Figs. 1 and 2 confirms the effectiveness of using the inductive approach in the
process of spectrum unfolding.
In conclusion, it should be added that for qualitative unfolding of a neutron energy
spectrum in the range 8-800 keV it is necessary to employ additional reactions, e.g., 37C1.
(n, y), LOS
Kh(n, n'), and 237Np(n, f), as well as 6Li(n, a), 10B(n, a), 295u(n,
f), and
239Pu(n, f) with irradiation in boron filters.
LITERATURE CITED
1. E. A. Kramer-Ageev, E. G. Tikhonov, and V. S. Troshin, Activation Methods of Neutron
Spectrometry [in Russian], Atomizdat, Moscow (1976).
2. C. Greer et al., A Technique for Unfolding Neutron Spectrum from Activation Measure-
ments, SC-RR-67-746 (1967).
3. A. Fisher and L. Turi, KFKI-71-22, Budapest (1971).
4. F. Rossitto and M. Terrani, Nucl. Instr. Methods, 79, 341 (1970).
5. J. Schmidt, KFK-120, Karlsruhe (1966).
6. B. Magurno, ENDF/B-IV Dosimetry File, BNL-NCS-50446 (1975).
7. A. A. Lapenas, in: Measurement of Neutron Spectra by the Activation Method [in Russian],
Zinatne, Riga (1975), p. 35.
8. M. Ya. Bankrashkova et al., At. Energ., 44, No. 3, 260 (1978).
110
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
KAL'MAR-1 PULSED ELECTRON ACCELERATOR WITH RELATIVISTIC
ELECTRON BEAM POWER OF UP TO 5?1012 W/cm2
B. A. Demidov, M. V. Ivkin, UDC 621.384.659
V. A. Petrov, and S. D. Fanchenko
- The possibility of achieving controlled thermonuclear fusion by using relativistic elec-
tron beams (REB), first pointed out by Zavoiskii [1]i is arousing ever-increasing interest.
As shown by Rudakov [2, 3], REB with a current of the order of 107 A and a power density of
the order of iO W/cm2 are required to accomplish this. The present paper describes the
Kai:mar 1 accelerator producing REB With a power density of 5,1011-541012 W/cm2. We give
the results of investigations on REB focusing in a high-voltage diode as a function of the
electrode geometryand the magnitude of the voltage prepulse. -
Accelerator Design. The design of the accelerator was chosen by Proceeding from the
following concepts. The Accelerator should have these parameters: electron energy 1 MeV,
impedance of high-voltage diode 1.5 SI, pulse duration 70-100 nsec. For normal operation of
the high-voltage diode the prepulse should not exceed 1-2% of the operating voltage. The circuit of
the accelerator should be as simple and reliable as possible. It was decided to use a duplex
shaping line (DSL) with water dielectric and one switching device (spark gap in water). The
4-0 output resistance of the DSL was matched to the impedance of the high-voltage diode by
using a coaxial tranformer with an output resistance of 1.5 0. One end of the central elec-
trode was electrically (and mechanically) connected directly to the central cylinder of the
DSL and the other end to the high-voltage diode (Fig. 1). The DSL (electrical length 70
nsec) is charged from a voltage-pulse (Arkad'ev--Marx) generator with an output voltage of
2 MV and a stored energy of 70 kJ. The generator, built of IMP-100-0,1 capacitors, consists
of 20 multiplier stages. The first stage uses a pressurized-gas trigatron-type spark gap
whereas the others have untriggered, open air spark gaps. Thanks to the insertion of an
inductance shunting the second-stage output to the ground, the scatter of the generator trig-
gering delay does not exceed Mt = 10-20 nsec when Tt = 250 nsec and the output voltage is
more than 80% of the breakdown voltage [8]. The DSL, the coaxial transformer 5 with an out-
put resistance of 1.5 S.2, and the high-voltage diode 7 are placed in a stainless steel case
(see Fig. 1) of diameter 1 m and length 4.2 m in which a forevaduum is set up during filling
with distilled water. The middle cylinder 3 of the DSL is mounted on stand-off insulators
with elastic elements to dampen pressure surge; these elements allow the cylinder to move
axially 3-5 mm.
The spark gap can operate in the spontaneous breakdown mode (the experimental data given
below were obtained with this mode) as well as in a laser-ignition mode. For laser ignition
of the water spark gap [5] the rod 2 was provided with a lens 12 to focus the laser radiation
onto the surface of electrode 3 and a protective Plexiglas window 11. The neodymium-glass
laser (pulse energy q,2 J, duration 20 nsec) consists of a driving oscillator 13 with a modu-
lated Q-factor and two amplifier stages 14.
The DSL charging circuit (Fig. 2) was chosen so that the prepulse amplitude could be
regulated over wide limits to within 1-2% of the amplitude of the main voltage pulse U without
an additional spark gap in the DSL-diode circuit. The capacitance of the generator (35 nF)
was chosen to be close to that of the DSL (capacitance of DSL branches: C2 = 14 nF, C9 =
12.5 nF), thus ensuring a high efficiency of resonant charging of the DSL (the charging period
corresponds to the cyclic frequency w = 1.7,106 sec-I). Since the resonant charging period
exceeds the pulse duration by an order of magnitude, if account is taken of the relations
between the values of the parameters of the electrical circuits, the prepulse amplitude can
be given by the approximate formula
AU= UL0C3(02,
(1)
Translated from Atomnaya fnergiya, Vol. 46, No. 2, pp. 100-104, February, 1979. Original
article submitted November 29, 1977.
0038-531X/79/4602-0111$07.50 ? 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
111
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
if
14
34
2 12 11
Fig. 1. Diagram of the apparatus.
Fig. 2
-HRH
13
5 10
6, mm
Fig. 3
Fig. 2. Equivalent circuit of apparatus in a charging mode: C1, C2, C2, C4)
capacitances of voltage-pulse generator, branch of DSL, and coaxial trans-
former, respectively; L) inductance of generator and charging circuit; Lo)
decoupling inductance; RI, R2, R2) resistance of leaks through water.
Fig. 3. Diode current (1, 2) and diode impedance (3, 4) as function of cath-
ode-anode gap: 0) prepulse amplitude 0.1 U; +) prepulse amplitude 0.01 U.
where Lo is the inductance of the high-voltage decoupling 10 (see Fig. 1) ensuring symmetric
charging of the DSL. When Lo = 2.2 pH the prepulse AU Pi 0.1 U,whereaswith Lo = 0.4 pH,
the prepulse AU Ps 0.01 U. In experiments without the laser the water spark gap operated in
the spontaneous breakdown mode. A nonlinear Willite resistance was introduced into the DSL
charging circuit to suppress postpulses in the generator-DSL circuit in cases when the spark
gap failed to operate or did so imperfectly [6]. The high-voltage diode was provided with
a plane anode 8 of diameter 80 mm and cathodes of different configurations were tried: plane,
needle, ring, convex cone, and hollow cone.
The current was measured with a noninductive resistance 6 and a Faraday cylinder. The
voltage across the DSL, across the input and output of the coaxial transformer, as well as
across the diode was measured with the voltage divider 4 described in [7]. In order to study
the REB self-focusing we used an x-ray camera obscura with a resolution of no worse than 0.1
mm. The Kal'mar-1 accelerator was equipped with an ion-exchange unit for desalinating water
(the total volume of distilled water was 3 m3 and the resistivity was 2,103 Q?cm).
112
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
'44
R,S2
5
W, Watt
1,8.10"
m.ww
/00 200t, asec
Fig. 4 Fig. 5
Fig. 4. DSL voltage amplitude at various cathode-anode gaps in di-
ode: a) IS = (Sic; b) IS < 6*; c) IS > (5*; d) calibration f = 1 MHz.
Fig. 5. Current and voltage oscillograms and dependence of diode
resistance and electron beam power on time and plane cathode geom-
etry.
Experimental Results. The first series of experiments on the Kalimar-1 was carried out
with a plane stainless steel cathode of diameter 70 mm. The experimental results are shown
in Fig. 3. The DSL voltage under these conditions was 0.5 MV. It is seen that in the case
of a large prepulse with gaps IS S 4 mm breakdown occurred in the diode so that the measured
current I differed little from the short-circuit current Isc.. It was possible, however,
.
(with both a small and a large prepulse) to choose a gap 6* at which the current I was roughly
one-half Is.c. , as should be the case with impedance matching of line and diode. The fact that
the line and diode impedances are matched is confirmed by the data from DSL voltage measure-
ments (Fig. 4), according to which the amplitude of the reflected voltage wave is a minimum
when IS = 6*. Figure 5 gives the experimental data on the current and voltage as well as on
the time dependence of the diode impedance and the output power of the accelerator for a
diode with a plane cathode. Under the operating conditions indicated, with the aid of an x-
ray camera obscura and by cleavages of material from the anode in some cases we observed self-
focusing of REB with a diameter of up to 2 mm.
Further experiments on the Kal'mar-1 accelerator were devoted to finding the conditions
for stable REB self-focusing at a given point on the anode. Different authors [8-10] sug-
gested that for this purpose use be made of a partially evaporating Plexiglas cathode at a
given moment by means of a laser or plasma injector. The search for the solution to the
problem of stable REB focusing was carried out on the Kal'mar-1 on the basis of changes in
the configuration of the metallic cathode. All of the cathode configurations enumerated
above were tried. The best REB focusing (focal point no greater than 1 mm in diameter) was
achieved by using a cathode in the form of a convex metallic cone with a base diameter of
70 mm and an angle of 135? (Fig. 6). X-ray camera-obscura photographs and cleavages of anode
material showed that such a cathode ensures REB focusing exactly in the center of the anode
with 100% duplication. The current density reaches reaches 107 A/cm2 and the power density,
5s1012 W/cm2. Good focusing is attained with a prepulse of both 0.01 U and 0.1 U. But be-
cause of the high impedance of the diode with a cathode in the form of a convex cone it was
113
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Fig. 6. X-ray camera-
obscura photograph.
0
100 200t nsec
Fig. 7. Current and voltage oscillo-
grams and time dependence of diode re-
sistance and electron beam power under
conditions of beam self-focusing.
not possible to achieve matching with the 1.5-Q output resistance of the line; as a result,
the absolute value of the energy contribution to the beam focus did not exceed 1 kJ.
The greatest energy contribution to the REB focal point was attained with a cathode in
the form of a hollow metallic cone of diameter 8 mm, first employed in the Angara-1 accelera-
tor [11]. Under these conditions the REB in the Kalimar-1 was focused at the center of the
anode, the current of the focused beam reached 200 kA, and the diameter of the focal point
was 2-3 mm. In addition to cleavages from the outside of the anode, formed in one pulse, the
REB formed a hemispherical crater of 5 to 5.5 mm deep on the inner surface of the aluminum
anode and in some cases pierced an anode made of 6-mm aluminum. According to computer cal-
culations [12], this corresponds to an energy of 3-4 kJ at the REB focal point. It must be
emphasized that such an energy contribution to the focal point is attained only under the
conditions of a small prepulse (0.01 U) at cathode?anode gaps of 1.5-3.5 mm in the diode
for voltages of 500 kV. The current and voltage oscillograms under the conditions of a large
energy contribution to the beam at the focal point (small prepulse) as well as the time de-
pendence of the diode impedance and the output power of the accelerator are shown in Fig. 7.
114
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Discussion of Results. The principal result is that the relatively simple circuit de-
sign of an electronic accelerator based on a DSL without peaking spark gaps makes it possible
to obtain REB with parameters which meet the requirements for performing experiments on the
construction of an electronic thermonuclear reactor. According to [13], a REB is absorbed
in heavy metals in a layer 5 to 10 urn thick, so that the energy release of 4 kJ at the focal
point of the beam in the Kal'mar-1 accelerator corresponds to an "instantaneous" energy re-
lease of '?,1 keV per atom of the material, just as in the proposed electronic thermonuclear
reactor.
The results of the measurements of the impedance R of the high-voltage diode with large
and small prepulses as well as with different cathode configurations are of interest in them-
selves. It is seen from Fig. 3 that the diode impedance is determined by two factors: the
concentration of the preliminary plasma and the length of the cathode?anode gap. With a
large prepulse and a given quantity 6 the diode impedance is almost twice as small as with a
small prepulse. The shape of curves 3 and 4 in Fig. 3 representing the cases of a small and
large prepulse in a plane-cathode diode are in good agreement with the relation R 62 stem-
ming from the three-halves power law. In both cases, however, the absolute value of R, as
in [14], differ from the value calculated from this law by a factor of several and depends
on the density of the preliminary plasma. It should be noted that with a large prepulse and
gaps of less than 5 mm the diode is quickly short-circuited by the plasma and short-circuit
conditions arise. In view of this, it is more convenient to operate with a small prepulse
admitting the use of small 6, thus making it possible to obtain high REB currents. When a
hollow-cone cathode was used with a small prepulse, we managed to reduce 6 to 2-3 mm and to
obtain a focused REB current of up to 200 kA. In this case the REB energy calculated by in-
tegrating the product of the measured diode current and voltage was 8 kJ, which is in satis-
factory agreement with the estimate of an energy contribution of 3-4 kJ from the damage of
the anode material.
Finally, it is of interest to note the time dependence of the diode resistance. It is
seen from the curves in Fig. 5 that the impedance of a plane-cathode diode changes insig-
nificantly in '1,100 nsec, as a result of which the diode impedance can be matched to the out-
put resistance of the coaxial transformer for a long period. In the focused-beam mode there
is some increase in the diode resistance with time (Fig. 7). This effect can apparently be
attributed to the anomalous resistance of plasma in turbulent processes associated with the
contraction of the beam into a narrow filament. The focused REB current of 200 kA at a volt-
age of 0.5 MV is in good agreement with the estimate from the formula proposed by Rudakov
for a relativistic beam [11],
/=0.3-17-U,
where I is the current in MA and U is the diode voltage in MV.
LITERATURE CITED
1. M. V. Babykin and E. K. Zaimovskyi, in: Plasma Physics and Controlled Nuclear Fusion
Research, IAEA, Vienna, Vol. 1 (1971), p. 635.
2. L. I. Rudakov and A. A. Samarskii, in: Proceedings of the Sixth European Conference on
Controlled Nuclear Fusion and Plasma Physics, Moscow, Vol. 1 (1973), p. 487.
3. I. P. Afonin et al., in: Proceedings of the Seventh Symposium on Engineering Problems
of Fusion Research, Knoxville (1973), p. 269.
4. B. A. Demidov and M. V. Ivkin, Prib. Tekh. Eksp., No. 3, 120 (1975).
5. B. A. Demidov et al., Prib. Tekh. Eksp., No. 1, 120 (1974).
6. B. A. Demidov et al., Prib. Tekh. Eksp., No. 3, 37 (1975).
7. B. A. Demidov and M. V. Ivkin, Prib. Tekh. Eksp., No. 2, 115 (1977).
8. V. I. Liksonov, Yu. L. Sidorov, and V. P. Smirnov, Pis'ma Zh. Eksp. Teor. Fiz., 19, 516
(1974).
9. Yu. V. Koba et al., in: Proceedings of the Fifth International Conference on Plasma
Physics and Controlled Nuclear Fusion, IAEA, Vienna, Vol. 2 (1975), p. 337.
10. P. Miller et al., Phys. Rev. Lett., 35, No. 14, 940 (1975).
11. M. V. Babykin et al., in: Technology of Intertial Confinement Experiments, IAEA, Vienna
(1977), p. 41 (Proc. IAEA Advisory Group Meeting, Dubna (1976)).
12. M. Widner and S. Thomson, Calculations of Anode Witness Plate Damage due to Pinched REB,
Sandia Lab. Report Sand-74-351 (1979).
115
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
13. S. L. Bogolyubskii et al., Pisima Zh. Eksp. Teor. Fiz., 24, 203 (1976).
14. G. A. Mesyats, Generation of High-Power Nanosecond Pulses [in Russian], Sovetskoe Radio,
Moscow (1974).
MEASUREMENTS OF DOSE EQUIVALENT OF MIXED RADIATION
OUTSIDE THE SERPUKHOV PROTON SYNCHROTRON SHIELD
A. V. Antipov, I. S. Baishev
V. T. Golovachik, G. I. Krupnyi,
V. N. Kustarev, V. N. Lebedev,
and M. Khefert
UDC 539.12.08
A comparison of the readings of radiation monitoring systems, in particular systems em-
ployed at various high-energy accelerators, is of practical interest both for the justifica-
tion of one or another method of measurement, and for further understanding and improvement
of the methods.
Comparative measurements of radiation doses outside the Serpukhov proton synchrotron
shield were made with the component method (CM) [1] and the "Cerberus" instrument [2, 3] used
at the Institute of High Energy Physics (THEP) and CERN as radiation monitors, and also with
the "Sukhona-2" recombination dosimeter described in [4], a linear energy transfer (LET) ra-
diation spectrometer based on a tissue equivalent (TE) proportional counter and a TE ioniza-
tion chamber, employed at the IHEP for the direct measurements of doses and for calibrating
monitoring devices used in radiation fields having an unknown composition and spectrum.
In the CM the dose equivalent is determined by using a rem-meter for neutrons with En <
20 MeV [5], a carbon activation 12C(x, xn)11C plastic scintillator detector to measure the
fluence of particles with E > 20 MeV, and an air-filled ionization chamber with aluminum
walls to determine the contribution of gamma rays and charged particles to the total dose.
A thermal-neutron detector in a moderator is used as a rem-meter for neutrons with Eh < 20
MeV [1].
The "Cerberus" instrument used as a radiation monitor at CERN in combination with a
12C(x, xn)IIC detector embodies another variant of the component method (Table 1). The in-
strument consists of a boron ionization chamber RIC in a cylindrical polyethylene moderator
[6], an aluminum ionization chamber AIR with a nonhydrogenous filler, and a tissue equivalent
ionization chamber TE.
Two algorithms are indicated for processing the results of the CM (THEP): the standard
[1] and one based on the response matrix [7], taking account of the depth distribution fac-
tors. The interpretation of measurements made by the CM (CERN) are described in detail in
[8]. All the instruments were calibrated with the same neutron and gamma sources under iden-
tical radiation geometry. The factor 35?10-9 rem.cm2 [9] was used to transform the fluence
of neutrons from a Pu--Be source to the dose equivalent.
Measurements were performed at three points outside the Serpukhov proton synchrotron
shield in the experimental room. Outside the upper accelerator shield the main contribution
to the dose was from hadrons with Eh > 20 MeV; in the vicinity of channel 11 neutrons with
En < 20 MeV predominated; near channel 2 high-energy muons made the main contribution.
To interpret the detector readings in terms of the maximum dose equivalent (MADE) [10]
the absorbed dose distribution in a phantom and the separate contributions to the dose made
by neutrons with energies up to 20 MeV and those above 20 MeV were studied. Figure 1 shows
measured absorbed doses and values calculated by integrating the dose distribution [11, 12]
over the neutron spectrum [13] in a cubical water phantom 30 cm on an edge irradiated at the
point of measurement outside the upper shield of the accelerator. The value of the dose in
the absence of the phantom at the point where its center is later located (the open circles
in Fig. 1) is taken as a unit. Figure 1 shows that the maximum dose in the phantom is 1.3
Translated from Atomnaya gnergiya, Vol. 46, No. 2, pp. 105-108, February, 1979. Original
article submitted February 20, 1978.
116 0038-531X/79/4602-0116$07.50 ? 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
TABLE 1. Configuration of IHEP and CERN
Component Methods; Algorithms for In-
terpreting Their Readings to Determine
the Dose Equivalent
IHEP
CERN
BF3 (SNMO-5),
noderator diam.
30 CM
-
E
--1
^-o
m
. ta
0 E
.1 0
tr) ID
.....".C1 ..Zt.
1,10
tt En
1
103Rh, moderator
diam. 25.4 cm
o
o
*0
al t.1
g Z
0 g
-
ii.
RIC
AIR
frl
f.
Ineutron rem-Meter
?.2
'c u -
"Cerberus" [8]
CM [1]:
-81 822 833
CM -2171: a2H+
-FasH3m; Ai-----s1jllr+e12H7+
+elan; A2 C2111T+822H7+
+823117; A3 = 831117'4'832n+
+833117
---=-HnH-Dy+HA;
Aortic Dy E,RnH n;
AAIR =-EDv+,nApn;
Ara_ +E.; ?
Hh=ChnFh
Note: H, max. dose equiv. (MADE) of.
mixed radiation; A, readings of detec-
tors; ci factors taking account of sen-
sitivity of detectors to various kinds
of radiation; 117, MADE of j-th compo-
nent of radiation; a, depth distribu-
tion factors; Hh, DT, Hh, Dh, doses of
radiation components; Fh, fluence of
particles with energies above 20 MeV;
Cling factor to convert fluence of neu-
trons with energies above 20 MeV to
dose equivalent.
times as large as the dose measured at this point without the phantom. This emphasizes the
importance of performing phantom measurements for the correct transformation to the MADE.
The difference between the experimental data and the calculated results arises from the fail-
ure to take complete account of the irradiation conditions in the calculational model: in
the first place the contribution to the dose from charged particles and gamma rays was not
taken into account, and secondly only the case of an infinite layer irradiated normally from
one or both sides was considered. Figure 2 shows the results of calculating the depth dis-
tribution of dose equivalents of neutrons with energies up to 20 MeV and those above, per-
formed for these same conditions and under the same assumptions.
Analysis of the phantom distributions of doses shows the importance of taking account of
their shapes in interpreting the results of the measurements. At the measuring point outside
the upper shield of the accelerator a correction factor of 1.3 must be applied to the readings
of the TE ionization chamber; the depth distribution factors for the MADE of neutrons with
energies up to 20 MeV and those above are 0.80 and 0.93, respectively.
Table 2 compares the measurements normalized to the same value of the absorbed dose. The
factor to convert the carbon detector readings to the equivalent neutron dose with Eh > 20
MeV was taken as 2.8010-8 rem?cm2 for all methods [14, 15]. The experimental results were
processed in the following way: CM-1 by the method of [1] (cf. Table 1); CM-2 by the method
in [7] using the response matrix; "Cerberus" by the method in [8]. At the point outside the
upper shield the depth penetration factors al = 0.80, a2 = 0.93, a3 = 1 were taken into ac-
count; the LET spectrometer readings were corrected by a factor of 1.3 to take account of
the absorbed dose distribution in the phantom (cf. Fig. 1).
Outside the upper shield of the accelerator where the information on the radiation field
is most complete and the MADE is most correctly determined from the point of view of the
117
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
JO
1,3
42.
441
:3440
2
138
20
'1'0 10 202 0
?
X, g. cm"
10 20 0.10 20 JO
g? cm2 Z, g ? cm-2
Fig. 1
? 10 20
Depth, cm
Fig. 2
JO
Fig. 1. Distribution of absorbed dose along the axes of a water phantom 30 x
30 x 30 cm irradiated at the point of measurement outside the upper shield of
the accelerator above the axis of the beam (0). The X axis coincides with the
direction of the beam and the.Z axis is directed upward: 0, absorbed dose in
the absence of phantom at the interseetion of XYZ axes (taken as a-unit). When
the phantom is in position its center is at this point;-----) calculated for
normal incidence of a broad beam of neutrons with the spectrum given in [13]
on a layer of tissue 30 em thick; ---) the same for irradiation from 'both
sides.
Fig. 2. Distribution of dose equivalent in a layer of tissue 30 cm thick ir-
radiated by a normally incident broad beam of neutrons, with an energy spec-
trum from [131: 1, 2) .Contributionof portions of spectrum with En < 20 and
En > 20 MeV respectively; 3) total distribution; HT + HT, MADE for portions
of spectrum; Hi and H2, contributions of portions of spectrum to total MADE
(H).
recommendations in [10] by-using the LET spectrometer, the results obtained are of consider-
able interest for the analysis of the divergences. It is clear that the use of the standard
algorithm CM-1 to interpret the CM at theTHEP leads to overestimates of the MADE reaching
70% for a rem-meter with an SNMO-5 counter in a 30-cm diameter moderator. The CM-2 component
method using the response matrix to interpret the detector readings and taking account of
the depth distribution factors of the separately detected radiation components at this point
gives a result very close to the MADE value of the dose equivalent. The results obtained
with'"Cerberus" are in somewhat worse agreement with the MADE. This results from the fact
that the algorithm for interpreting the detector readings [8] does not take account of all
the corrections introduced in the case of CM-2, for example the contribution of particles
with energies above 20 MeV to the ionization chamber readings [5, 8]. Analysis of the re-
combination dosimeter and LET spectrometer differences is complicated by the large differ-
ences in detector geometries.
For measurements in the channel 11 area the recombination dosimeter readings were taken
as a unit, since in the 0.2-20-MeV range the mean energy of the spectrum of neutrons which
make the predominant contribution to the dose was estimated as 0.5 MeV by using the method
in [16]. It is improper to use the LET spectrometer in such a neutron spectrum since in this
case the particles do not produce a triangular pulse spectrum in the volume of the counter
[18-20]. These differences in the values of the total dose are accounted for by the behavior
of the response functions of the neutron rem-meters for En < 20 MeV in the energy range be-
low 0.5 MeV [3, 21, 22] where the functions are considerably above the MADE curve for neutrons
recommended by the ICRP [23].
118
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
TABLE 2. Comparison of Measurements Outside Shield of Serpukhov Proton Synchro-
tron
Place of
measure-
ment
En*,
MeV
MethodsH
to-,orem
D,
to-ad
-C1
HT 1 HT
HT
H/HLET
10-6 rem
Upper shield
(hadrons)
5
M-1 (SNMO-5)
33,6+2,0
10,5+0,7
3,2+0,6
13,2+1,3
15,4+1,5,
5,0+0,5
1,71+0,20
M-1 (SNM-3)
29,5+1,8
10,0+0,7
3,0+0,3
9,1+0,9
15,4+1,5
5,0+0,5
1,51+0,18
M-2(SNM-3)
24,0+2,0
7,0+0,4
3,4+0,4
9,6+1,0
14,0+1,6
3,3+0,8
1,22+0,16
"Cerberus"
26,9+1,6
6,4+0,3
4,2+0,3
9,2+0,6
15,2+1,5
2,5+0,3
1,37+0,16
"Sukhona - 2"
28,6+3,0
6,8+0,3
4,2+0,4
1,46+0,21
LET spectrometer
19,6+2,0
7,0+0,4
2,8+0,3
T
Channel 11
(neutrons)
0,5
M-1 (SNMO-5)
38,7+3,1
9,9+0,7
3,9+0,3
26,0+3,0
8,2+0,8
4,5+0,5
2,15+0,32
M-1 (SNM-3)
21,1+1,2
7,6+0,6
2,8+0,3
8,4+0,8
8,2+0,8
4,5+0,5
1,17+0,16
M-2 (SNM-3)
19,1+1,4
5,4+0,3
3,5+0,4
8,0+0,8
6,9+0,9
4,2+0,8
1,06+0,16
"Cerberus"
29,3+1,5
4,8+0,3
6,1+0,5
17,5+1,1
8,2+0,8
3,6+0,4
1,63+0,22
"Sukhona- 2''
18,0+2,3
5,0+0,3
3,6+0.4
1
LET spectrometer
10,8+1,1
5,4+0,3
2,0+0,2
0,60+0,10
Channel 2
(muons)
1
M-1 (103Rh)
6,7+0,6
5,9+0,6
1,1+0,2
0,6+0,1
0,43+0,04
5,7+0,6
1,03+0,13
CM-2t ('Rh)
6,6+0,9
5,4+0,3
1,2+0,2
0,6+0,1
0,20+0,30
5,9+0,9
1,02+0,17
"Cerberus"
5,3+0,3
5,2+0,3
1,0+0,1
0,43+0,10
0,43+0,04
4,4+0,3
0,82+0,09
"Sukhona -2"
6,3+0,3
LET spectrometer
6,5+0,6
5,4+0,3
1,2+0,1
Note: H, total dose equivalent; Dm, absorbed dose; Q, average quality factor
of mixed radiation; HT, HT, MADE of neutrons with En < 20 and > 20 MeV respec-
tively; HT, MADE of gamma rays and charged particles. The relative errors of
the measurements are indicated everywhere.
*En, average energy of fast neutrons (0.2-20 MeV) estimated by method in [16].
tUsing the response matrix at this point in the processing, the contribution
to the dose from high-energy muons was determined from the charged-particle
counter [17] and the dose characteristics of muons from [12].
At channel 2 where more than 90% of the dose is due to high-energy muons, the results
of all methods agree within the limits of error of the measurements. The differences in the
values of the gamma and charged-particle doses obtained by using the CM-1 and "Cerberus" are
accounted for by the difference in thickness of the walls of the aluminum chambers used.
Analysis shows that the main reasons for the divergences of the results in determining
the dose equivalent by component methods in radiation fields with a preponderance of high-
energy hadrons are the differences in interpretation algorithms; in radiation fields with a
soft neutron spectrum the divergences result mainly from differences in the response func-
tions of the rem-meters used. In both cases it is important to take account of the shape
of the depth dose distributions of the separately detected radiation components and the dif-
ferences in radiation spectra and depth dose distributions in calibration and in measurements.
Values obtained by the CM may differ from the results of TE detectors because of differences
in the principles of operation; the response functions of TE detectors model the dependence
of the quality factor on the LET of the radiation, while the response functions of the com-
ponent method detectors model the dependence of the MADE of particles of a definite type and
energy range on their energy.
Thus, by the example of the interpretation of readings of component method detectors
using the response matrix it has been shown possible to eliminate the main systematic errors
without changing the instrumental part of the method. By using the matrix it is possible to
determine the MADE of mixed radiation with an error of no more than 30% in radiation fields
with a preponderance of hadrons with Eh > 20 MeV, 10% with a preponderance of neutrons with
energies En < 20 MeV, and 5% with a preponderance of high-energy muons, which corresponds to
the recommendations of the ICRU for practical radiation monitoring, and ensures agreement of
the results of monitoring at various accelerator complexes.
119
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
In conclusion, the authors thank M. N. Chimankov for help in performing the measure-
ments, and all the co-workers of the Department of Radiation Research of the IHEP for taking
part in the discussion of this work and for their remarks and helpful advice.
LITERATURE CITED
1. V. E. Borodin et al., Preprint IHEP 74-131, Serpukhov (1974).
2. M. Hbfert, Preprint CERN DI/HP/187, Geneva (1975).
3. M. 'Wert, Preprint CERN DI/HP/177, Geneva (1974).
4. V. T. Golovachik et al., in: Proceedings of the International Congress on Protection
against Accelerator and Space Radiation, CERN, Geneva, Apr. 26-30, 1971. CERN 71-16,
Vol. 1, p. 431.
5. I. S. Baishev et al., in: Proceedings of the Fifth All-Union Conference on Charged
Particle Accelerators [in Russian], Vol. 1, Nauka, Moscow (1977), p. 224; Preprint CERN
HS-RP/104, Geneva (1977).
6. I. Anderson and J. Brown, Nucleonics, 6, 237 (1964).
7. V. T. Golovachik et al., Preprint IHEP 77-114, Serpukhov (1977).
8. M. H8fert and M. Nielsen, Preprint CERN HP-75-142, Geneva (1975).
9. D. Nachtigall, Health Physics, 13, 213 (1967).
10. E. B. Keirim-Markus (Ed.), Radiation Safety. Papers 19 and 20 ICRU. Atomizdat, Moscow
(1974).
11. K. Shaw, G. Stevenson, and R. Thomas, Preprint RHEL/M149, Chilton (1968).
12. V. T. Golovachik et al., in: Proceedings of the Fourth All-Union Conference on Charged
Particle Accelerators [in Russian], Vol. 2, Nauka, Moscow (1975), p. 212.
13. E. A. Belogorlov and V. S. Lukanin, ibid., p. 236.
14. M. Hofert and J. Baarli, ibid., p. 207.
15. V. T. Golovachik, V. N. Kustarev, and V. N. Lebedev, Preprint IHEP 77-91, Serpukhov
(1977).
16. D. Hankins, in: Proceedings of the IAEA Symposium on Neutron Detection and Standardiza-
tion, Vol. 2, IAEA, Vienna (1963), p. 123.
17. A. M. Biskupchuk, S. A. Drugachenok, and G. I. Krupnyi, Preprint IHEP 76-120, Serpukhov
(1976).
18. R. Dvorak, Health Phys., 17, 279 (1969).
19. H. Rossi and W. Rosenzweig, Radiology, 64, No. 3, 404 (1955).
20. M. M. Komochkov, Soobshchenie OIYaI R16-10647, Dubna (1977).
21. Robert S. Sanna, Preprint HASL-267, N. Y. (1973).
22. E. A. Kramer-Ageev, in: Problems of Dosimetry and Radiation Shielding [in Russian],
No. 6, Atomizdat, Moscow (1967), p. 3.
23. ICRP Publication 21. Data for Protection against Ionizating Radiation from External
Sources: Supplement to ICRP Publication 15. Pergamon, Oxford (1973).
120
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
OBITUARY
VIKTOR MIKHAILOVICH GUSEV
B. B. Kadomtsev, V. V. Orlov,
M. S. Ioffe, Yu. V. Martynenko,
V. V. Titov, and O. B. Firsov
Viktor Mikhailovich Gusev, one of the leading workers of the I. V. Kurchatov Institute
of Atomic Energy, an eminent specialist in atomic physics and solid-state physics, and head
of the laboratory of the Plasma Physics Division, died on Oct. 6, 1978.
V. M. Gusev was born in 1919. In October 1941 he volunteered for the front while a
third-year student in the physics department of Moscow State University. Before going into
action, he joined the Communist Party of the Soviet Union. He was seriously wounded during
fighting near Leningrad but after a long convalescence he returned to duty and fought until
victory. He was decorated with the Order of the Great Patriotic War and many medals. When
the war ended he returned to Moscow State University and graduated.
Over the more than 30 years of his scientific career, Viktor Mikhailovich Gusev made a
significant contribution to a number of areas of physics. In the early years of his career
he participated actively in the solution of the problems of creating atomic engineering in
the country. He was awarded a State Prize for this work. A cycle of studies was done under
his leadership in 1958-1960 on the implantation of deuterium ions in metals (by means of the
D(d, n)3He reaction) as well as on measurement of the coefficients of metal sputtering by
deuterium ions. These investigations served as the beginning of the study of the influence
of wall material on the processes of obtaining and confining high-temperature plasma.
V. M. Gusev began working in the I. V. Kurchatov Institute of Atomic Energy in 1960.
In 1961 he came out with the initiative for the creation of a new field of science and en-
gineering, viz., ion implantation of semiconductors. Under his leadership and with his di-
rect participation this field underwent an extensive development. He did a large complex of
work, ranging from formulation of problems and elaboration of the physical foundations of the
Translated from Atomnaya gnergiya, Vol. 46, No. 2, pp. 109-110, February, 1979.
0038-531X/79/4602-0121$07.50 0 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
121
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
method of ion implantation to the practical realization of scientific and technological de-
velopments. His scheme and design were used to construct the small ILU high-current ion ac-
celerator with mass separation of ions, making it possible to obtain currents of 300-keV
doubly and triply charged ions of sufficient magnitude for practical use. Commercial 4:.toduc-
tion of ILU units was started in 1966. These accelerators have been set up in many scien-
tific and industrial organizations of our country as well as in Hungary, the German Democratic
Republic, and Bulgaria. Initially, this ensured the broad development of work on ion im-
plantation, the introduction of a new progressive method into the fabrication of semiconductor
devices based on silicon, especially photoelectric solar energy converters, p--i?n commuta-
tion diodes, diodes with various purposes, MOS and bipolar transistors, and integrated cir-
cuits. Along with his co-workers, V. M. Gusev was the first to obtain a p--n structure with
ion beams in an ILU accelerator; this structure is characterized by high injection properties
and a record-breaking intensity of radiation in the visible region of the spectrum.
V. M. Gusev made a great contribution to the development of the physical foundations of
ion implantation: he studied the spatial distributions of implanted ions in amorphous and
crystalline targets, the law of the generation and annealing of defects in implantation lay-
ers, the effect of ion channeling and radiation-stimulated diffusion on the spatial distribu-
tions of implanted ions as well as on the electrical properties of these layers.
Viktor Mikhailovich possessed exceptional generosity which was manifested in his sharing
his experience with others. Workers of many leading institutes of our country as well as
Hungary and the German Democratic Republic have trained in the laboratory he headed. The
staff of the TsIFI Ion Implantation Laboratory, headed by I. Dyulai, regards V. M. Gusev as
one of those scientists because of which work on ion implantation in semiconductors in Hungary
was initiated and developed extensively.
V. M. Gusev used his experience on ion implantation to create radiation-resistant coat-
ings on metals. Investigations, which V. M. Gusev first began, on chemical reactions during
the interaction of hydrogen ion beams with various materials have taken on great importance.
In 1974 the scientific interests of Viktor Mikhailovich became linked with a new sub-
ject, i.e., working out materials science problems of thermonuclear fusion. He employed the
ILU accelerator successfully to simulate the process of interaction of plasma with the first
wall of a thermonuclear reactor.
Viktor Mikhailovich recently devoted particular attention to problems pertaining to the
processes of synergism under the conditions of large-scale thermonuclear reactors. With his
active participation work began on the creation of new devices with which it will be possible
to simulate the concurrent action on the wall of various components of the corpuscular radia-
tion of plasma and to imitate the effect of fast neutrons on the materials. Realizing that
the conditions in large tokamaks are the closest to those in future reactors, he significantly
facilitated the construction of a diagnostic complex for studying the processes of plasma
interaction with the wall now being built in the T-10 tokamak.
The untimely death of V. M. Gusev is a great loss for Soviet physics. His bright image
will remain in the minds of all who knew him.
122
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
LETTERS
DETERMINATION OF THE ABSOLUTE YIELDS OF 43.5-, 74.7-, AND
117.8-keV y PHOTONS FROM 243AM
Yu. S. Popov, D. I. Starozhukov,
V. B. Mishenev, P. A. Privalova,
and A. I. Mishchenko
UDC 539.163
The isotopic composition of our 24'Am sample, as determined by mass spectrometry, was
99.476 ? 0.019% 2'Am and 0.524 ? 0.019% 241Am. The half-lives of 243,Am and 24 Am were taken
as T2/2 = 7370 ? 40 yr and 432.7 ? 0.7 years, respectively [2].
The americium was purified of plutonium, neptunium, and their decay products in a 1 x 8
Dowex anion-exchange column (25-50 min, column dimensions 3.5 x 10 cm). Pu (IV) and Np (IV)
were sorbed successively from a nitric acid solution (8 moles/liter). The plutonium was sta-
bilized in the IV-valence state by heating in nitric acid (8 moles/liter) with hydrogen perox-
ide. Np (V) was reduced to Np (IV) by hydrazine nitrate in HNO3 (8 moles/liter) by heating
in a water bath for 30 min.
Radiochemically pure americium in the form of a nitric acid solution with no more than
a 2% content of inert admixtures by weight was deposited in a layer '1,0.6 pg/cm2 thick on a
'1,110 pm pg/cm2) polyethylene backing. No more than 2 h elapsed from the instant purifi-
cation was stopped to the end of the measurements. During.this time the 239Np activityreached
less than 2% of its equilibrium value.
The y spectrometer included a coaxial 30-cm3 Ge (Li) detector", an SS-03 spectrometer
assembly, and a buffer register with output to a BSM-4M computer. The spectrometer ensured
an energy resolution of 2.5 keV at half.-height of the total absorption peak of 5Co at 122
keV.- The spectrometer was calibrated with two sets of standard spectrometric y sources.
The values of the counting efficiencies were calculated from the nuclear physics con-
stants used earlier in [3]. The specific a activity of the 243AM solution was determined by
absolute 4Tra, counting. The values of the absolute y yield were calculated as the weighted
mean of several series of measurements.
The errors of the absolute values of the y yields were determined mainly by the errors
of the areas under the total absorption peaks (2-20%)?, the' errors in determining the absolute
a activity of 24'Am in the sample (4-5%), and the error in the y counting efficiency (4-9%).
These errors amounted to 4,22, '?,7, and '1,14%, respectively, for 43.5-, 74.7-, and 117.8- keV y
TABLE 1. y Yields of 243AM at Various
Energies
Quantum yield, quanta/decay, ob
Literature
43,5 keV
74,7 keV
117,8 keV
4+1
69+3
? 0,5
[4]
5,3*
61-*
0,75
[5]
5+1
73+3
0,64+0,25
[6]
5,5+0,3
66+3
[11
100 t
0,84t
[8]
59,1+4,0
13]
5,3+1,2
60+4
0,7+0,1
*The authors of [5] did not cite the ab-
solute errors of the values indicated.
tRelative measurements.
Translated from Atomnaya gnergiya, Vol. 46, No. 2, p. 111, February, 1979. Original
article submitted February 2, 1978.
0038-531X/79/4602-0123$07.50 ? 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
123
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
rays. The values cited for the relative errors correspond to a confidence coefficient of
0.95.
Our results are compared with other published data in Table 1. The values obtained for
the quantum yields agree with data in [3, 5, 7].
LITERATURE CITED
1. L. Brown and R. Propst, J. Inorg. NucI. Chem., 30, 2591 (1968).
F- Oetting and 8. Gunn, J. Inorg. Nucl. Chet., 29, 2659 (1967).
3, D. I. StarOzhukov, Yu. S. Popov, and P,,A. Privalova, At. Energ., 42, No, 4, 319 -(1977):
F. Asaro -et Al.,-Phys. ReV,, 117, 492 :(1960)-
J, Van Hise. and D Engelkemeir, Phys..Rev., 171, 1325 (1968)
6. B. M. Aleksandrov, P. I, Grigoriev, and N S. Shimanekaya, Yad. Fiz,, 10, 14 .(1969).
7 J Ahmed and M. Wahlgret, Nuci Instrum. Methods, 99, 333 (1972).
8., .Z ..=Pate et. Phys, A272, 169 (1975)
ONE MICROWAVE METHOD OF DOSIMETRY FOR PULSES OF
PENETRATING RADIATION
Kapiaos and Yu. A. MedvedeV
In an earlier paper [1] we considered the possibility of Using microwave methods for
the dosimetry of powerful short pulses of penetrating radiation in an air of quite low den-
sity when the electron lifetime 6 Is considerably longer than the pulse duration T. In
another -paper 121 we considered possible microwave methods of dosimetry in air of high den-
sity when T >> 0. In this case one microwave method is used to measure the electronic con-
ductivity G(t) of the air, which is in a linear relation with the exposure dose M(t) with
lO6M C;10" R?sec-':
if(0==WG(0,sec-1*.
The time dependence of the exposure dose rate can be determined by measuring any quan-
tity which is proportional to the electronic component of the conductivity of the ionized
air. Such a quantity which can be used conveniently is the damping of probing electromag-
netic waves during their passage through a transmission line filled with ionized air:
a (0 == PI OP (0,
where Pi(t) and P(t) are the intensities of the probing electromagnetic signal at the output
of the transmission line with ionized and un-ionized air, respectively.
Using the data of [5], we can show that at sufficiently high electron collision rates
veff:> w, where w is the cyclic velocity of the probing field, the wave intensity at the out-
put of a transmission line uniformly filled with a conducting medium is related to the con-
ductivity G(t) of that medium by
4aL
(t)= I ln 1?a (t) I = G (t),
where L is the length of the transmission line. From this we find that for 106 < M < 101' R?
sec-', when the ionic component of the conductivity of the air can be neglected in comparison
with the electronic component and the process of electron-ion recombination, for low damping
a 1 we have:
137c
111 (t) = 47tr CC (1) ?
(1)
For experimental verification of the correspondence between the transient shape of the
damping pulse (and, therefore, of the pulse of the exposure rate) and the transient shape of
*We used the values given in [3, 4] as constants.
Translated from Atomnaya gnergiya, Vol. 46, No. 2, pp. 112-113, February, 1979. Original
article submitted February 22, 1978.
124 0038-531X/79/4602-0124$07.50 C) 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Fig. 1
Fig. 1. Oscillogram of pulse of exposure
diation. Scanning rate 1 division/11sec.
Fig. 2. Oscillogram of electron current pulse with energy of
8 MeV. Scanning rate 1 division/usec.
46-,A
41
Fig. 2
rate of electron ra-
0
o
D-
0
0
? 0 ?
1
2 t, ?sec
Fig. 3. Time dependence of ratio of
electron current I(t) to damping a(t)
of rf signal.
the pulse of ionizing radiation we simultaneously recorded the exposure rate pulse (Fig. 1)
with a measuring instrument based on a strip line [6] with a length of 3.46 m, operating at
a frequency of 900 MHz, and a pulse of electronic current of amplitude 31 mA (Fig. 2) with
an energy of the order of 8 MeV by means of a Faraday cylinder. Comparison of the oscillo-
grams shows that the shapes of the exposure rate and electron pulses are practically the same
(Fig. 3). It follows from Fig. 3 that during the electron pulse the ratio I(t)/a(t) does
not deviate from the value 0.31 by more than 5% (a somewhat larger deviation at the end of
the pulse when tl.8 usec was due to insufficiently ?high time characteristics of the-Fara-
day-cylinder method). The results given in Fig. 3 experimentally confirm the validity of
relations of the type of Eq. (1),
Af (0 == Ka (0, (2)
where K is a time-independent coefficient which can be calculated as follows. The exposure
rate and the number Q(t) of electron pairs formed per cm' of air in 1 sec are related by
M (t) (t)12,08.109 ,
and Q(t) can be written as
v 1 (t)
Q (0=7 AO"
S
(3)
(4)
where I(t) is the time dependence of the electron-current pulse (in A), S is the area of the
cross section of the Faraday cylinder (cm2), and v/Z = 40 cm-1 [7] is the number of secondary
electrons formed in air per unit length by Compton electrons with a mean energy of the order
of 1 MeV.
When Eqs. (2)-(4) are taken into account,
v I (t)
K =3.3.109 S1 a (t)
125
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
TABLE 1. Comparison of Results of rf
and Thermoluminescence Dosimetry Meth-
ods
Mmax ? 1 0-7,
R ? sec-1
Mmax.t -1 0-.1
R?sec-1
Mmax,thAmax
0,84
1,6
1,9
0,92
1,9
2,1
1,6 .
2,6
1,6
2,1
3
1,4
3,6
6,6
1,8
11
15
1,4
12
22
1,8
20
27.
1,4
40
48
1,2
'
i.e., if it is noted that I(t)/a(t) = 0.31 A and S = 40 cm2, we get K = 140 c/4nL which is
in complete agreement with Eq. (1).
In a number of experiments, along with measurements of the exposure rate by the rf method
we measured the radiation dose Dt by the thermoluminescence method with VA-S-220 dosimeters
(see Table 1). In the process, the relation
M =M
max, t max
Di
S M (t)dt
where Mmax, the maximum value of the exposure rate M(t) recorded by the rf method in the .
same experiment as Dt was recorded, was used to convert the radiation dose Dt integrated over
a pulse into the maximum dose rate Mmax,t obtained by the thermoluminescence method.
The fact that the ratio 1..1-
-max,t/Mmax is constant within the limits of the error of measure-
ment (the error of measurement is 15% in the case of the rf method and 25% in the case of the
thermoluminescence method, the mean value of the ratio is 1.6) for a whole train of pulses
attests to the applicability of the rf method for measuring the exposure rate of ionizing ra-
diation,whereasthe systematic deviation of these ratios from unity is due to the difference
in the method of calibration of the rf and thermoluminescence dosimeters.
LITERATURE CITED
1. Yu. A. Medvedev et al., At. Energ., 40, No. 1, 53 (1976).
2. Yu. A. Medvedev, N. N. Morozov, and B. M. Stepanov, At. .frierg., 45, No. 5, 374 (1978).
3. Yu. A. Medvedev, B. M. Stepanov, and G. V. Fedorovich, in: Problems of the Metrology
of Ionizing Radiation [in Russian], Atomizdat, Moscow (1975), p. 183.
4. V. N. Kapinos et al., Zh. Tekh. Fiz., 44, No. 11, 2432 (1974).
5. V. Golant, Microwave Methods of Studying Plasma [in Russian], Nauka, Moscow (1968), p.
296.
6. A. L. Feltdshtein, L. R. Yavich, and V. P. Smirnov, Handbook of the Fundamentals of
Waveguide Engineering [in Russian], Sovet-skoe Radio, Moscow (1967), p. 214.
7. L. A. Artsimovich, Handbook of Nuclear Physics [in Russian], Fizmatgiz, Moscow (1963),
p. 361.
ERRATA
In the article "Liberation of Hydrogen from Aqueous Solutions Irradiated in Nuclear Reactors"
by M. V. Vladimirova and I. A. Kulikov (Vol. 45, No. 3, 1978) the formula on p. 927 should
read:
126
I +f.n=2.5' 105 (pt.t4 eViml?sec.
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
USE OF THERMAL-NEUTRON PROBES TO MEASURE
THERMAL-NEUTRON FLUX OF DISTRIBUTIONS
Yu. A. Saf in, S. G. Karpechko,
P. G. Afanas'ev, V. I. Nalivaev,
V. B. Pampura, and V. I. Uvarov
UDC 621.039.51
At the present time considerable attention is being paid to the monitoring of energy
release in the cores of nuclear reactors for various purposes [1]. Exact and reliable mea-
surement of the neutron fields in reactor cores requires continuous improvement of old ex-
perimental methods and development of new ones. In-reactor measurements, however, always
entail certain difficulties due to the difficult radiation and thermal conditions under which
the detectors must operate, etc. In recent years there has been a great development of in-
strumental methods of measuring neutron fields, methods in which small fission chambers, elec-
tron-emission neutron, self-powered probes, thermal-neutron probes, etc. are used as the
primary probes [2-5].
In the present paper we consider the possibility of employing specially developed ther-
mal-neutron probes for operational measurement of the thermal-neutron flux distribution over
the height and radius of the core of the IVV-2 research reactor [6]. A TND-A measuring probe
was designed and built for this purpose. It consisted of two small thermal-neutron sensing
elements: a high-sensitivity (HS) and a low-sensitivity (LS) sensing element [3610-16 and
3610-16 V/(neutrons/cm26sec), respectively], placed in an aluminum can with a diameter of 54
mm (the diameter of the working part of the probe was 18 mm). The upper part of the can con-
tains a device for moving the probe over the height of the reactor core in 50-mm steps. The
design of the thermal-neutron sensing elements is much like that described in [4, 5]. The
transmission line in the sensing element consists of a KTMS high-temperature cable with two
aluminum thermoelectrodes. The signal is measured by a PP-63 potentiometer with recording
on an EPP-09 M3 potentiometer.
The TND-A probe was installed in one cell of the reactor core and put in the lowest
position, after which measurements were made. The probe was then raised 50 mm, fixed in
that position, and the signal was measured once again. Measurements were made in succes-
sion over the entire height of the cell and the probe was then moved to the next cell. All
the measurements were carried out at a reactor power of 200 kW and a water temperature of
35 ? 5?C. The reactor power and the position of the control devices (automatic regulator
and shim rods) during the measurements were kept constant. The thermal-neutron flux dis-
tribution was measured in 24 reactor core cells: six water "traps," nine fuel assemblies,
and nine beryllium slugs. A 0.9-mm copper wire was the indicator. The wire was fastened
in special holders made of acrylic plastic. The irradiation was carried out for 10 min at
a reactor power of 10 kW. The counting system consisted of a plastic scintillator of the
BG type, an FEU-19M photomultiplier, an amplifier, a power supply, and a PST-100 scaler. The
position of the reactor control devices during the measurements of the distribution by the
activation method was kept the same as in the measurements with the TND-A probe. The re-
sults of the measurements are given in Table 1 and Fig. 1.
It is seen from Table 1 that the signal from the HS sensing elements is almost an order
of magnitude greater than that from the LS sensing element whereas the relative distribu-
tions of the thermal neutron flux, obtained by these elements, coincide with good accuracy
in all the cells of the reactor core. This warrants the conclusion that the LS sensing ele-
ment can be used to monitor the density of thermal-neutron fluxes neutrons/cm26sec.
The relative distributions of the copper-wire activity and the readings of the TND-A probe
with thermal-neutron sensing elements in the beryllium slug and in the fuel assembly are of
the same character and coincide within the limits of error of measurement. In the water
trap the results differ somewhat. This difference can be attributed to the different spectral
Translated from Atomnaya gnergiya, Vol. 46, No. 2, pp. 114-115, February; 1979. Original
article submitted March 6, 1978.
0038-531X/79/4602-0127$07.50 ?1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
127
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
,o
0,5
7: 10
48
46
,e7
0,4
0 70 20 30 40 I-1, c m
Fig. 1. Relative distribution of thermal-neutron
flux, measured with TND-A probe and by ac-
tivating a copper wire (---) at various distances
H from channel bottom.
TABLE 1. Signals from HS and LS Sen-
sing Elements in Some Characteristic
Cells of Reactor Core, mV
Distance from
endofprobe to
bottom ofre-
actorccme,
cm
Fuel as-
se mbly
Water trap
Beryllium
slug
HS
LS
HS
LS
HS
LS
0
4,79
0,55
8,98
1,10
6,13
0,72
5
5,29
0,62
11,90
1,45
6,85
0,80
10
5,94
0,69
14,0
1,72
7,94
0,95
15
6,42
0,72
15,26
1,90
8,80
1,05
20
6,15
0,72
15,45
1,92
8,85
1,07
25
5,56
0,66
14,10
1,74
8,66
1,05
30
4,98
0,58
12,30
1,48
7,83
0,94
35
4,13
0,50
10,14
1,22
6,69
0,80
40
3,19
0,42
7,85
0,94
5,35
0,66
45
2,68
0,34
5,09
0,65
4,15
0,53
50
2,62
0,34
3,37
0,44
3,52
0,48
sensitivities of the copper indicator and the thermal-neutron probe as well as inadequate cen-
tering of the probe in the water trap (the coefficient of nonuniformity of the thermal-neutron
flux distribution over the radius of the water trap is about 1.2). A similar effect was ob-
served in [7].
The results permit the recommendation of probes with thermal-neutron sensing elements
for operational in-reactor monitoring of the thermal-neutron flux distribution in the range
from 5.1010 to about 1014 neutrons/cm2esec.
LITERATURE CITED
1. I. Ya. Emel'yanov et al., Control and Safety of Nuclear Power Reactors [in Russian],
Atomizdat, Moscow (1975).
2. A. B. Dmitriev and E. K. Malyshev, Neutron Ionization Chambers for Reactor Engineering
[in Russian], Atomizdat, Moscow (1975).
3. V. B. Klimentov, G. A. Kolchinskii, and V. V. Frunze, Activation Measurements of Neutron
Fluxes and Neutron Spectra in Nuclear Reactors [in Russian], Standarty, Moscow (1974).
4. I. Ya. Emel'yanov et al., At. Energ., 30, No. 3, 275 (1971).
5. J. Boland, Monitoring Instruments for Nuclear Reactors [Russian translation], Atomizdat,
Moscow (1973).
6. A. P. Zyryanov et al., in: Radiation Safety and Protection of Atomic Power Plants [in
Russian], No. 2, Atomizdat.,, Moscow (1976), p. 110.
7. S. S. Lomakin et al., At. Energ., 30, No. 3, 301 (1971).
128
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
TWO-DIMENSIONAL MODELING OF THE FUEL ASSEMBLIES OF
HIGH-TEMPERATURE GAS-COOLED REACTORS
M. D. Segal' and V. I. Khripunov UDC 621.039.517.5
High-temperature gas-cooled nuclear reactors (HTGR) [1] are becoming very important in
power engineering and for producing high-potential heat. Such reactors are characterized by
relatively high calorific intensity combined with long-term resources and strict require-
ments on the limiting fuel temperature. In this connection it is quite important to equalize
the temperature fields of the fuel elements and the working substance, especially in the case
of cluster-type reactors.
As a result of the slug effect and perturbations contributed by the regulating elements,
etc., spatial nonuniformities can develop in the heat release field of the fuel assemblies
(consisting of a large number of fuel-element rods) of such reactors. This nonuniformity
gives rise to local overheating of the working substance and the fuel and therefore causes a
reduction in the parameters of the fuel assemblies. A way to raise the output parameters of
such assemblies that appears promising is the two-dimensional modeling of the energy-release
field of fuel concentrations. Here we are of course speaking only of zone modelling, since
it evidently would be technically impossible to individually choose concentrations of fuel
for each fuel element (the Gulf General Atomic (U.S.A.) reactor uses assemblies containing
270 fuel elements). Physical modeling methods were discussed by Vinterberg [2] and further
developed and experimentally justified by others [3-5]. It has been shown that changes in
the concentration and the fuel-to-moderator volume ratio may possibly have a considerable
effect on the distribution of energy-release sources in various systems. In many cases, the
condition that the temperature at the surface or the center of a fuel element is kept con-
stant reduces to an equalization of the density of energy release over the volume.
The optimization of these parameters with respect to the reactor as a whole is based on
certain simplifying assumptions (a one-dimensional geometry, the few-group approach, etc.).
In modeling a fuel assembly, it is also necessary to take into account the mixing of the
working substance and the local characteristics of the energy distribution. The basis of the
present approach consists in simultaneously using a multidimensional Monte Carlo neutron-
physics calculation [6] and a heat-hydraulic calculation [7] to enable a rather effective
equalization of the temperature fields of the fuel elements and the gas coolant. The tem-
perature fields were calculated in three dimensions for a homogeneous heat-releasing medium,
using a BgSM-6 computer and a program which has been previously discussed [8].
In order to fix the spatial energy distribution in the inhomogeneous medium within the
framework of the Monte Carlo method, a local estimate of the flux in extended (filament)
mathematical detectors was proposed. For each collision of a neutron within the volume of the
system, the contribution to the energy release is proportional to the integral
q (1/A) exp ? lix2+ A2)/1(x2 4_ Az dr,
0
(1)
where q is the contribution to the energy release due to the scattering of the neutron at a
point, A is the optical distance between the detector and the point at which the neutron is
scattered, and X is the optical length of the detector. In a correlated calculation for a
large number of detectors, certain properties of the estimate (a single-parameter dependence
on A which is due to rapid convergence with respect to the upper limit) make it possible to
reduce the error in the local energy-release estimate in an r,T geometry down to 3-5%. A
relationship for the fuel concentration in the modeling zones was selected on the basis of
the equality of the average values of the energy release in each zone. The constraint on the
maximum allowable fuel temperature was represented by the equation
Translated from Atomnaya fnergiya, Vol. 46, No. 2, pp. 115-117, February, 1979. Original
article submitted March 13, 1978.
0038-531X/79/4602-0129$07.50 0 1979 Plenum Publishing Corporation 129
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
36 9gK
950
Fig. 2
Fig, 1. 'Lines of constant energy release (in arbitrary units) for
an dnmodeled fuel aSsembly.
Fig.. 2. Isotherms of the gas at the exit of an unmodeled assembly;
Fig. 3. Modeling zones.
si
Fig. 4. Isotherms of the gas at
the exit of a modeled assembly.
Ci (E)Oi (r, ?E) dE dSit S (Di (r,T,E)dE dSi=const,
Si E
(2)
where Si is the area of the modeling zone, E is the energy of the neutrons, Ci is the fuel
concentration in the modeling zone, and (Di is the neutron flux in the modeling zone.
It should be noted that in the case of assemblies where the energy release fields have
high gradients, it is desirable to model the fields on the basis of the equality of the maxi-
mum values of the energy release in the modeling zones. The choice of the number and config-
urations of the modeling zones is extremely important. Using a linear programming method de-
scribed in [9], the zone configurations were chosen with respect to a certain energy-release
field for a fixed number of fuel concentrations in a fuel element. The coefficient of non-
uniformity of the energy release over the cross section of the modeling zones was minimized
in the process. The zone configurations and the relation for the fuel concentration in them
were finally corrected, taking into account the hydrodynamics and the local characteristics
of the energy distribution.
The variational theory proposed in [10] can be applied to the problem of estimating the
energy release to show that the correction to the exact value due to a change in the configu-
ration of the modeling zone is of second order. Only two iterations were needed in anumerical
130
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
experiment to keep the error in the energy release distribution from exceeding 2-3%.
As an illustration of this method, we shall consider the modeling of a fuel channel with
a hypothetical energy release field.
Characteristics of the channel
Power, MW
1.5
Helium:
consumption, kg/sec
0.85
intake temperature, ?C
300
Number of fuel elements in a cassette
270
Diameter of the fuel element, mm
7.2
The hypothetical field of energy release is shown in Fig. 1, and the corresponding iso-
therms of the gas at the exit are shown in Fig: 2. Three fuel concentrations were used in
modeling this channel. The configuration of the modeling zones is shown in Fig. 3. Figure
4 shows the isotherms of the gas at the exit which correspond to the new energy release field.
The fuel concentrations in the zones are in the ratio C1/C11/C111 = 0.848/1/1.19. The non-
uniformity of the exit gas temperature field thus decreased from 1.27 to 1.13 (with respect
to the heating). This made it possible to increase the power of the fuel assembly by almost
10%.
It should be noted that the proposed method eliminates the radial and azimuthal non-
uniformity of the energy release. In so doing, the number of fuel elements with different
concentrations of fuel is no larger than in the case of radial modeling.
LITERATURE CITED
1. D. Bedenig, Gas-Cooled High-Temperature Reactors [Russian translation], Atomizdat, Mos-
cow (1975).
2. F. Vinterberg, in: Transactions of Second Geneva Conference, FRG Proceedings [in Rus-
sian], No. 1055, Atomizdat, Moscow (1959), p. 453.
3. N. N. Ponomarev-Stepnoi and E. S. Glushkov, At. gnerg., 11, No. 1, 19 (1961); 12, No.
5, 415 (1962); 20, No. 6, 478 (1966).
4. N. N. Ponomarev-Stepnoi, E. S. Glushkov, V. I. Nosov, and S. N. Barkov, ibid., 28, No.
1, 58 (1970).
5. A. P. Rudik, Optimal Distribution of Nuclear Fuel [in Russian], Atomizdat, Moscow (1974).
6. D. V. Markovskii, V. I. Khripunov, and G. E. Shatalov, A Multigroup Calculation of
Heterogeneous Reactors by the Monte Carlo Method [in Russian], Institute of Atomic En-
ergy Preprint No. 2959, Moscow (1977).
7. M. D. Segal', Calculation of the Temperature Fields in an rITiz Geometry in Porous Media
with a Spatial Nonuniformity of Heat Release [in Russian], Institute of Atomic Energy
Preprint No. 2845, Moscow (1977).
8. M. D. Segal', A Brief Description and Instructions for the Use of a Program for Cal-
culating Temperature and Velocity Fields [in Russian], Institute of Atomic Energy Pre-
print No. 2785, Moscow (1977).
9. C. Tzanos, Trans. Am. Nucl. Soc., 24, 195 (1976).
10. G. I. Bell and S. Glasstone, Nuclear Reactor Theory, Van Nos Reinhold (1970).
131
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
MAXIMUM RATE OF EMISSION OF LONG-LIVED y-EMITTING
AEROSOLS ALLOWABLE UNDER ATOMIC POWER STATION
STANDARDS AND CONTROL
G. G. Doroshenko, E. S. Leonov,
Z. E. Lyapina, V. A. Fedorov,
and K. N. Shlyagin
UDC 543.52+614.73
In order to determine the radial distribution of "Co activity in the soil near an
NVAgS, natural measurements were taken at radial distances of 0.5, 1, 2, 3, and 5 km; the
values obtained for the specific activity of "Co were 5, 4, 2, 1, and 0.5 mCi/km2, respec-
tively. Its total activity within a radius of 5 km was estimated by integration to be 97
mCi [1]. This value is in agreement with the result (113 mCi) obtained from measurements of
the total emission during a 10-yr period of operation of the NVAgS after taking into account
the physical decay of the "Co.
Measurements of layered samples were used to estimate the distribution of "Co with re-
spect to depth in the earth. It was approximately exponential, with a characteristic inverse
length K = 0.5 cm-1. These data could be used to calculate the exposure dose rate at a dis-
tance h = 1 m from the surface of the earth, using the equation [2]
P (E0, h, K)=1.44.103Eovf (E0, h, K) F, (1)
where E0 is the photon energy in MeV; y, electron conversion factor in cm-1; v, quantum yield
in photons/decay; F, specific activity in mCi/km2; and f, a tabulated function.
The values of dose rate P calculated for the above points are equal to 0.86, 0.69, 0.34,
0.17, and 0.086 mrd/yr, which are less than 1% of the natural radiation background for this
region [3].
On the basis of the experimental data, and with the assumption that the distribution of
specific activity F(R) remains the same for various values of the accumulated activity Q0, a
nomogram was constructed (see Fig. 1) in which the daily emission q is connected with the
equilibrium quantity of total activity by the equation
Qo =---qTi/2/1n 2.
(2)
It follows from the nomogram that in case the emission is due solely to "CO, an allow-
able rate of emission of long-lived aerosols of 0.5 Ci/day [4] leads to unjustifiably high
dosages in the districts close to the atomic electric power station. Realistic emission
levels from an NVAgS correspond to 0.04 mCi/day and the maximum dose rate in the locality is
at most 1 mrd/yr at a distance of 0.5 km from the atomic electric power plant. On the basis
of this, it is proposed that the allowable operating level of "Co emission rate from an
atomic power plant be set such that its equilibrium activity should not exceed 100 mCi; i.e.,
the quantity accumulated in the environment near the NVAL. Since it is difficult to measure
a small allowable emission rate and is unnecessary for operative control, it is proposed in-
stead of daily to set up a monthly allowable "Co emission rate of 1.2 mCi/month for a water-
modulated?water-cooled power reactor with a total output of 1455 MW (electric) (Which scales
to 0.9 mCi/month for 1000 MW (electric)).
It is not possible to study other similar long-lived nuclides because of the smallness
of the quantities being measured against the background of natural activity and global fall-
out. The maximum yearly emission of "41137Cs, 141,144ce, 103,106Ru, Mn, "Zr, "Nb, "Cr,
and 12?mAg for four NVAgS units amounts to ,v460 mCi/yr or 1.26 mCi/day, which is 400 times
as small as the maximum allowable emission. Assuming that the distribution of long-lived
nuclides in the neighborhood of an atomic power plant is no different from the measured "Co
distribution, and taking the maximum emission of each of the above radionuclides into account,
Translated from Atomnaya Energiya, Vol. 46, No. 2, pp. 117-118, February, 1979. Original
article submitted March 15, 1978.
132 0038-531X/79/4602-0132$07.50 CI 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
q.mCiAlaY
Fig. 1. A nomogram constructed according to' data
obtained from measurement's on a NVAgS. a, b, c, d) -
exposure dosages of 5, 0.5, 0.17, 0.005 rd/yr, re-
spectively; 1, 2, 3, 4, 5) distance from the atomic
energy power plant of 0.5, 1, 2, 4, 5 km; emission
tate, .mCi/day: -----) 0.04; --- 1; x- x) 500; 0)
experimental results.
we estimated the exposure dosage rate at a distance of 0.5 km from an NVAgS. The calculation
was carried out for an equilibrium activity of each radionuclide accumulated in the environ-
ment using Eqs. (1) and (2), with a result of 7.5 mrd/yr..
We see in this way that even when the activity level in the environment reaches equilib-
rium, including 137Cs (T2/2 = 30 yr), the exposure dosage rate near an atomic energy power
plant is at most 5% of the natural background and is 0.5% at a distance of 5 km from the
atomic energy power plant in a calculation assuming an output level of 1000 MW (electric).
A rate of emission of long-lived y-emitting aerosols of 460 mCi/yr or 300 mCi/yr for an out-
put of 1000 MW (electric) may be taken as an approximate figure for purposes of setting a
standard for the appropriate operating level of allowable emission rate.
LITERATURE CITED
1. G. G. Doroshenko, E. S. Leonov, K. N. Shlyagin, and V. A. Fedorov, in: Proceedings of
the COMECON Scientific--Technical Conference, Prague (1976), pp. 24-31.
2. G. A. Fedorov, I. E. Konstantinov, and V. V. Pavlov, in: Dosimetry and Radiation Shield-
ing [in Russian], No. 7, Moscow (1967), pp. 87-98.
3. E. I. Vorob'ev et al., At. Lerg., 43, No. 3, 374 (1977).
4. Safety Rules for Atomic Electric Power Plant Design [in Russian], Ministry of Public
Health of the USSR, Moscow (1969).
133
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
A CONTACTLESS METHOD OF STUDYING THE THERMAL STATE
OF FUEL ELEMENTS DURING IRRADIATION
V. N. Murashov, L. S. Kokorev, UDC 621.039.517.5
and V. V. Yakovlev
PreVioU;vatudieSjI741:haveshown that the,ther0Ophysical properties of fuel elements
can undergo considerable change during irradiation. The results of Asatato 15] and MacDonald
and Weisman 16]Andicatelthat during the initial period of irradiation (100-300 h) thetem-
perature at the center Of the fuel elements rises, and then drops during the stbsequentpe-
riod Of operation at a constant output level.
,Tbe'authOrt:Of the 'present study propose a Method of contactiess COntrolpf_thethermal
state of fuel elements during a4srOcess Of prolonged operation. The method is based on the
use of non?tationary regimes with intermittent changes of output, interruptions, actuations
of the emergency protection Of the reactor, and fluctuations of output A-neutron flux de-
tector at the TVS: and a thermocouple in the coolant flux at the exit of the active tone serve
as detecting elements.:. A relation between the perturbations Of the coolant temperature (u)
and the perturbations Of the linear thermal load of the TVS (.S) is obtained by solving the
system of energy equations for the coolant and fuel element. The solution is obtained in the
form
?
u (t) = exp ( ? az) u (0, 0) exp[-nt (t?ft)]-1/- 1z1(t--0) 1, 12 Vi z Al'
+M exp [ ? a (z?)] S 0)exp [?m (t?)110 [21i/ (t-0) I dft ca,
0
where m = 11/(CR), R is the integral coefficient of the thermal resistance between the fuel
and the coolant (including the thermal resistance of the fuel, the contact clearance between
fuel element and jacket, and the convection heat exchange), C is the heat capacity per unit
length of the fuel element in the TVS, II is the perimeter of heat exchange of the fuel ele-
ments in the TVS, a = HIRGziCzj, Gzi and Czi are the mass consumption and heat capacity of
the coolant, 1.. = am, M = m/GzzCzi, and Io and I are Bessel functions of imaginary argument.
It follows by definition that m is proportional to the integral thermal conductivity
from the fuel to the coolant.
The required parameter here is m = 1/T, i.e., the reciprocal of the temperature field
period of relaxation T. The relationship between the relaxation period and the thermophysical
properties of the fuel element was found in [4]. They found the unknown parameter by mini-
mizing the mean squared deviations during time T between the calculated (u) and experimental
(ue) relations for the perturbations of coolant temperature at the TVS exit, using Eq. (1)
to calculate u. The method of least squares can be used to obtain and equation for determin-
ing m:
(ue? u) cdurn dt 0,
(2)
which was solved on a computer.
The experimental measurements were made on an assembly with fuel elements of the water-
moderated-water-cooled power reactor type, man MR reactor. A TVS with 18 fuel elements
was equipped with thermocouples at the entrance and exit and a thermal neutron detector so
that the variations in the energy release could be observed. Figure 1 shows the results of
Translated from Atomnaya fnergiya, Vol. 46, No. 2, pp. 118-120, February, 1979. Original
article submitted March 27, 1978.
134 0038-531X/79/4602-0134$07.50 C) 1979 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
sec
44
43
42
_r
n_
JT-
10 100 450 ? 750 850 00 00 450 1600 HO 2;50 2350 280 ROO 450
Time,b
Fig. 1. Variation of the relaxation period (x) and of the linear thermal load
averaged over the TVS q1 (solid curve) during irradiation.
300
300
sec-
0,30
0,25
0 700 200
Time, II
Fig. 2
300
400
50 150 250 fe, w /cm
Fig. 3
0
Fig. 2. Variation of the period of relaxation of the temperature
field during the initial stages of the irradiation.
Fig. 3. Dependence of the period of relaxation on the linear ther-
mal load of the fuel element. 1) starting dependence; 2) 850 h of
irradiation; 3) 2150 h of irradiation; 4) calculated curve.
the measurements of m during a period of over 4500 h from the start of the irradiation. Dur-
ing the operation the output was repeatedly changed and the emergency protection actuated
(indicated by the arrows). As indicated by the experimental data, the thermal conductivity
of a fuel element drops during the first 100 h but subsequently increases (Fig. 2). The data
shown are for a linear thermal fuel element load averaged over the TVS qi 1, 200W/cm. Actua-
tion of the emergency protection after q,380 h results in a sharp reduction in the conductivity,
and with subsequent operation at constant output the conductivity increases to a certain
stable level. It follows from Fig. 3 that during operation the period of relaxation drops
by a factor q,1.4, corresponding to a temperature drop at the center of a WMWR-1000 type fuel
element from 2100 to 1600?C for a thermal load of 500 W/cm. Under prolonged operation in
the high thermal load region, the fuel element reaches a state in which the thermal resistance
of the fuel-jacket contact becomes negligibly small in comparison with the total thermal re-
sistance (curve 4 in Fig. 3). These results are in qualitative agreement with the data on
the behavior of oxide fuel [1-6].
LITERATURE CITED
1. B. Lastman, Radiation Phenomena in Uranium Dioxide [Russian translation], Atomizdat,
Moscow (1964).
135
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
Declassified and Approved For Release 2013/02/12 : CIA-RDP10-02196R000800010002-4
2. I. G. Lebedev et al., in: Problems of Atomic Science and Engineering. Series "Radia-
tion Materials Science, Methods and Techniques of Irradiation" [in Russian], No. 6,
Dimitrovgrad (1975), P. 19.
3. Kh. Khoffman, At. Tekh. Rubezhom, No. 4, 19 (1976).
4. V. N. Murashov et al., A Design Calculation and Experimental Study of the Center Tem-
peratures in Uranium Dioxide Fuel Elements, Institute of Atomic Energy Preprint No.
2936, Moscow (1978).
5. R. Asamoto et al., Center Temperature Measurements of Mixed Oxide Fuel from Zero to
3'105 MWd/Te, GEAP-13603 (1970).
6. P. MacDonald and J. Weisman, Nucl. Technol., 31, No. 3, 357 (1976).
DISTRIBUTION OF TRITIUM IN TECHNOLOGICAL SYSTEMS OF
THE NOVOVORONEZH NUCLEAR POWER PLANT
D. P. Broder, L. I. Golubev,
V. M. Ilyasov, A. I. Luee,
B. N. Mekhedov, I. R. Nurislamov,
L. N. Sukhotin, L. P. Kham'yanov,
and V. M. Arkhipkin
UDC 621.039.524.44
The radioactive hydrogen isotope tritium presents a definite radioactive danger to nu-
clear power plant personnel and to the population. It is formed as a result of the activa-
tion of the coolant and impurities in it, and also by ternary fission of the nuclear fuel.
The principal state of tritium in water coolant is HTO, which does not differ chemically
from ordinary water and cannot be separated by special water purification devices. There-
fore as a result of leaks in the primary loop tritium is freely propagated through the tech-
nological systems of the power plant and escapes into the service rooms and into the environ-
ment. In rooms with tanks of pure distillate for the primary loop where the total radioac-
tivity (without tritium) is below 3.10-10 Ci/liter, the concentration of tritium can reach
10-5 Ci/liter. Therefore practically the total dose to personnel coming into close contact
with this water is determined by tritium.
To estimate tritium doses and emission during a run it is necessary to know the tritium
concentration in the technological loops of the power plant. To find this concentration we
used a mathematical model which takes account of the connection of the loops of the third and
fourth units of the Novovoronezh Nuclear Power Plant with the feedwater and discharge sys-
tem.
For steady-state water exchange (Fig. 1) the dilution ratio of the coolant in the feed-
water system by uncompensated water can be assumed constant. The concentration of tritium
in the coolant of the primary loop can be calculated by solving the tritium balance equation
in the technological system of the primary loop:
k 1_ tT 114---fX-Ho (1-01C,
dt ? Vp
where C is the concentration of tritium in the coolant, E0 is the macroscopic cross section
for the formation of tritium in the reaction 1513(n, 2a)5H at the beginning of the run, E04)
is the reaction cross section averaged over energy, T is the time of a run, E0[1 ? (t/T)] cor-
responds to the decrease in the concentration of boron in the coolant to zero at the end of
the run, Vp is the volume of coolant in the primary loop of the third and fourth units, Ve
is the volume of coolant in the core, X = 1.53010-4/day is the tritium decay constant, w =
(3.9 ? 0.7)*10-5/day is the fraction of the coolant removed per day from the primary loop
system, c = 0.05 ? 0.01 is the dilution ratio of the coolant in the feedwater system by un-
compensated water. Since A