SOVIET ATOMIC ENERGY VOLUME 17, NO. 6

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP10-02196R000600120002-4
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 148
• Document Creation Date: December 23, 2016
• Document Release Date: March 14, 2013
• Sequence Number: 2
• Publication Date: December 1, 1964
• Content Type: REPORT

A
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Volume 17 No. 6
,
rc
December, 1964
SOVIET
ATOMIC
ENERGY
ATOMHAFI 3HEprifa
(ATOMNAYA iNERGIYA)
TRANSLATED FROM RUSSIAN
CONSULTANTS BUREAU
q?)
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
PLENUM PRESS HANDBOOKS OF
HIGH-TEMPERATURE MATERIALS
No. 1 ? MATERIALS INDEX ,
by Peter T. B. Shaffer Foreword by Dr. Henry H. Hausner
Data on the general, chemical, electrical, mechanical, nuclear, optical, structural, and
thermal properties of approximately 520 refractory material are arranged under
each material. Materials discussed fall into the categories of borides, oxides, carbides,
nitrides, silicides, mixed carbides, mixed Oxides and elements.
,
All compounds for which information was available are included. The concise listing
of property data under each individual refractory compound will greatly facilitate
retrieval of both general and specific data on the properties of an individual material.
The work is di,stinguished by a bibliography of 690 references to the literature on
refractory compounds.
782 pages $17.50
No. 2? PROPERTIES INDEX
by G. V. Samsonov _ Foreword by Dr. Henry H. Hausner
Revised by the author to include references to investigations as recent as late 1963.
Extensive tabular data is presented to provide a scientific classification of 600
refractory compounds, systematically arranged by the following properties: crystal-
chemicab thermal, thermochemical, electrical, magnetic, optical, mechanical, chem-
ical, and refractory properties of borides, carbides, nitrides, silicides, phosphides, and
sulfides of metals, as well as nitrides, carbides, and phosphides of boron and silicon
and boron-silicon alloys. Invaluable information is also presented on the principal
fields of application of refractory compounds in the metallurgical, chemical and
maChine construction industries, in power generation, and in automation, radio, and
electrical engineering. An appendix gives the most up-tozdate phase diagrams 'of
systems in which refractory compounds are formed.
-
The work contains a bibliography of over, 1300 references to the literature on refractory
compounds, approximately 40% of which are to Soviet literature hitherto compara-
tively undocumented in the West.
430 pages Translated from Russian $22.50
No. 3 ? THERMAL RADIATIVE PROPERTIES
by W. D. Wood, H. W. Deem, and C. F. Lucks
A compilation of data on thermal radiative properties originally published in two
soft-cover volumes by the Defense Metals Information Center, Battelle Memorial
Institute. The importance of the data for all those concerned with radiant heat
transfer' has prompted republication as a part of this significa?nt reference series.
Thermal radiative data are included for the following materials: titanium and its
alloys; stainless steels; iron-;nickel-, and cobalt-base superalloys; refractory metals
(chromium, columbium, molybdenum, tantalum, and tungsten) and their alloys;
coated materials for elevated-temperature service; ceramics; graphite.
476 pages $17.50
Contents on request
cP PLENUM PRESS 227 W. 17th St., New York, N.Y. 10011
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
A KOALA A A 2,P4511.1..11
EDITORIAL BOARD
A. I. Alikhanov
A. A. Bochvar
N. A. DollezhaP
K. E. Erglis
V. S. Fursov
I. N. Golovin
V. F. Kalinin
N. A. Kolokollsov
(Assistant Editor)
A. K: Krasin
I. F. Kvartskhava
A. V. Lebedinskii
A. I. Leipunskii
M. G. Meshcheryakov
M. D. Millionshchikov
(Editor-in-Chief)
1.1. Novikov
V. B. Shevchenko
A. P. Vinogradov
N. A. Vlasov
(Assistant Editor)
M. V. Yakutovich
A. P. Zeilrov
SOVIET ATOMIC
ENERGY
A translation of ATOMNAYA iNERGIYA
A publication of the Academy of Sciences of the USSR
? 1966 CONSULTANTS BUREAU ENTERPRISES, INC.
227 West 17th Street, New York, N.Y. 10011
Vol. 17, No. 6
December, 1964
CONTENTS
Some Ways of Development for Water-Moderated Water-Cooled Power Reactors
?A. Ya. Kramerov, Yu. V. Markov, S. A. Skvortsov, V. P. Denisov, E. V. Kulikov,
Yu. P. Sorokin, V. V. Stekornikov, A. A. Khokhlachev, V. P. Tatarnikov,
and V. A. Sidorenko
The Organic-Cooled Organic-Moderated Nuclear Power Station "Arbus"?K. K. Polushkin,
I. Ya. Emel'yanov, P. A. Delens, N. V. Zvonov, Yu. I. A leksenko, I. I. Grozdov,
S. P. Kuznetsov, A. P. Sirotkin, Yu. I. Tokarev, K. P. Lavrovskii, A. M. Brodskii,
A. R. Belov, E. V. Borisyuk, V. M. Gryazev, V. D. Tetyukov, D. N. Popov,
Yu. I. Koryakin, A. G. Filippov, K. V. Petrochuk, V. D. Khoroshavin, N. P. Savinov,
M. N. Meshcheryakov, V. P. Pushkarev, V. A. Suroegin, P. A. Gavrilov,
L. N. Podlazov, and I. N. Pogozhkin
TS-3 Compact Atomic Power Station?N. M. Sinev, A. K. Krasin, I. F. Bychkov,
0. I. Blokhin, D. L. Broder, V. N. Gabrusev, Yu. V. Dudnikov, V. A. Zhil'tsov,
M. A. Koptev, A. Ya. Komarov, A. P. Kotov, M. N. Lantsov, G. A. Lisochkin,
G. A. Merzlikin, I. G. Morozov, Yu. I. Orekhov, Yu. A. Sergeev, P. N. Slyusarev,
G. N. Ushakov, N. V. Fedorov, V. Ya. Chernyi, and V. M. Shmelev
Physical and Operating Characteristics of the SM-2 Reactor?S. M. Feinberg, N. A. Dollezhal',
E. D. Vorob'ev, V. A. Tsykanov, I. Ya. Emeryanov, V. M. Gryazev, A. S. Kochenov,
Yu. M. Bulkin, V. I. Ageenkov, and P. G. Aver'yanov
The PGR Pulsed Graphite Reactor?I. V. Kurchatov, S. M. Feinberg, N. A. Dollezhal',
P. I. Aleshchenkov, F. S. Drozdov, I. Ya. Emeryanov, A. D. Zhirnov,
M. A. Kazachenko, G. D. Knyazeva, F. V. Kondrat'ev, V. D. Lavrenikov,
N. G. Morgunov, B. V. Petunin, V. P. Smirnov, V. M. Talyzin, A. G. Filippov,
I. L. Chikhladze, P. M. Chulkov, and Ya. V. Shevelev
Statistical Reactor Kinetics Equations?A. B. Govorkov
Channel Effects in Fission of Even-Even Compound Nuclei?L. N. Usachev, V. A. Pavlinchuk,
and N. S. Rabotnov
Neutron Angular and Energy Distribution at the Boundary of Two Media?V. A. Dulin,
V. G. Dvukhsherstnov, Yu. A. Kazanskii, and I. V. Shugar
The Neutron Background at the Surface of the Earth?G. V. Gorshkov, V. A. Zyabkin,
and 0. S. Tsvetkov
Addition of Hetero-Organic Compounds to Polystyrene?E. E. Baroni, S. F. Kilin,
T. N. Lebsadze, I. M. Rozman, and V. M. Shoniya
PAGE
ENG. I RUSS.
1183
427
1197
439
1207
448
1212
452
1224
463
1236
474
1242
479
1249
486
1256
492
1261
497
Annual Subscription: $95
Single Issue: $30
Single Article: $15
All rights reserved. No article contained herein may be reproduced for any purpose what-
soever without permission of the publisher. Permission may be obtained from Consultants
Bureau Enterprises, Inc., 227 West 17th Street. New York City, United States of America.
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
CONTENTS (continued)
PAGE
EN6,
RUSS.
In .Metnory of Konstantin Konstantinovich Aglintsev
1265
591
LETTERS TO THE EDITOR
Study of the Transformations.of Ruthenium Dioxide in the Presence of Chromium Oxide
?M. K. Baranaev, V. G. Vereskunov, and K. P. Zakharova
1267
502
Use of .the Time of Flight Method for Measuring the Range/Energy Relation for 18 to 38 MeV
Helium IbriS in Aluminum?N. I. Venikov and N. I. Chumakov ?
1269
503
Charge-Exchange of Oxygen Ions of Energy 2-13.3 MeV in Thin Alundum Films
?N. I. Venikov, N. I. Chumakov, and B. I. Khoroshchavin
1271
504
Neutron Radiative Capture in Copper and Molybdenum?V. A. Tolskikov, V. E. Kolesov,
A. G. Dovbenko, and Yu. Ya. Stavisskii
1272
505
Radiative Capture Cross Sections for Fast Neutrons in Iron?A. V. Malyshev,
Yu. Ya. Stavisskiii and A. V. Shapar'.
1277
508
Use of the Monte Carlo Method for Calculating the Penetration of y- Radiation Through
Matter?L. M. Shirkin
1279
509
SCIENCE AND ENGINEERING NEWS
Symposium on the Biological Effects of Radioisotopes
1282
512
Special-Purpose Heavy Cement with Enhanced Absorbing Power?K. S. Kutateladze
and A. V. Rustannbekov
1286
515
BIBLIOGRAPHY
New Books
1288
517
Author Index, 1964
1295
Tables of Contents
1300
The Russian date "Podpisano k pechati" of this issue was 11/17/64 . This is equivalent to "approved
for printing." Publication did not occur prior to this date, but must be assumed to have taken place reasonably
soon thereafter.
Publisher
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
SOME WAYS OF DEVELOPMENT
FOR WATER-MODERATED WATER-COOLED POWER REACTORS*
A. Ya. Kramerov, Yu. V. Markov, S. A. Skvortsov, V. P. Denisov,
E. V. Kulikov, Yu. P. Sorokin, V. V. Stekol' nikov, A. A. Khokhlachev,
V. P. Tatarnikov, and V. A. Sidorenko
Translated from Atomnaya nergiya, Vol. 17, No. 6,
pp. 427-438. December, 1964
FEATURES OF REACTOR CONSTRUCTION AND SCHEME FOR THE SECOND BLOCK
OF THE NOVO-VORONEZH NUCLEAR POWER STATION
Water-moderated water-cooled power reactors using water as moderator and heat carrier are represented in
the USSR by two blocks of the Novo-Voronezh Nuclear Power Station (NPS), of which the first block, with 210 MW
electric power, is just starting operation and the second (365 MW electric power) is being constructed.
The first block of the station was described earlier [1-3]. In constructing this, the aim was to gather experi-
ence in planning, setting up, and operating stations using as heat source heterogeneous reactors with water under
pressure. Naturally in this first large NPS the parameters and planning decisions were effected with some degree of
caution.
The aim in planning the second Novo-Voronezh NPS was, on the one hand, to take a step forward in the crea-
tion of atomic-power systems with economic indices approximating those of thermal power plants, and, on the other,
to make full use of the experience in planning and setting up the basic equipment gained when constructing the first
station, while avoiding new departures, the realization of which would demand a considerable consumption of time
and resources on scientific -research and constructional work. Hence the development of the second block may be
regarded as limited, by the framework indicated, to the modernization of the first station.
- -? As in the first block, a body with dimensions limited by transport conditions will be used. This will be de-
void of anticorrosion plating, which will allow the thickness of the main metal to be increased and the calculated
pressure to be raised to 120 ohs. atm. From considerations of unification, the active zone, as in the first block, is
made up of 349 hexagonal cassettes of gage 144 mm and length 2.5 mm. The fuel elements are cylindrical, with
sintered uranium dioxide in shells, ? 0.5 mm thick, of zirconium-niobium alloy or stainless steel. To shorten the
installation time it was decided to use the master-circulation glandless pumps adopted by industry and used in the
first block (discharge 5250 m3/11, head 60 m water). For the same reason, recharging the fuel with removal of the
roof was retained; at the same time a system for recharging without roof removal was developed so that this might
be used in future.
In view of the fact that the removal and replacement of the roof, the actual fuel charging, and the restoration
of the reactor system to the operating state could occupy a great deal of time, these operations should preferably be
effected as rarely as possible, i. e., the length of operation between two rechargings should be fairly prolonged. For
increased specific power of the reactor this means not only a corresponding increase in the depth of combustion but
at the same time a rise in the initial store of reactivity, which must be compensated by absorption in the control
and safety elements of the reactor.
The principle of compensating for excess reactivity based on the replacement of fuel by a fast-neutron water
trap is used. The structural work showed that the number of such elements could be raised to 73. These are placed
uniformly about the active zone at the nodes of a triangular lattice with a period of 294 mm.
*Report No. 304 presented by the USSR to the Third International Conference on the Peaceful Use of Atomic En-
ergy, Geneva, 1964.
1183
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
The water/uranium ratio (ratio of the volume of water to that of the fuel material) is kept the same as in the
first block (1.7). This ensures a slight negative coefficient of reactivity, sufficient compensating power of the con-
trol system, and a good neutron balance.
The fuel is loaded into cassettes of roughly the same construction and with the same cross-sectional dimen-
sions as in the first block. For a core diameter of 7.7 mm and an external diameter of the fuel elements 8.8 mm,
each cassette holds 127 fuel elements in a triangular lattice with spacing 12.3 mm. In this way it proved possible
to bring the thermal power of the reactor to ? 1400 MW, i. e., 1.8 times greater than that of the first block. A
further substantial increase in power could only be realized by increasing the consumption of water and revising
the principles of the reactivity compensation and loading system, which would contradict the original intentions
of the design.
In a compaign lasting ? 1.5-2 y, with three partial rechargings, a burn-up of ? 22,000-30,000 MW ? day/ton
uranium is achieved. The required average enrichment for the first charging is 2.5-3.5%; the feed of the reactor
is maintained by 3-4% enriched fuel.
Out of the 73 control-element cassettes, 60 (CR) are intended for the compensation of slow changes in re-
activity (control rods) and 13 (SR) for emergency protection (scram rods). The construction of the CR and SR is in
principle analogous to that of the cassette in the first-block reactor. Provision is made, however, for hydraulically
relieving them from upward stress from the flow of heat carrier, which in this more thermally-stressed reactor pre-
vents reliable downward motion of the cassette (in the sense of reduced reactivity).
The SR cassette, just as that of the CR, consists of two parts: the absorber and the part containing fuel. Use
of fuel in the SR makes it possible to improve the neutron balance in the active zone and raise the depth ofburn-up.
It was decided to use a saturated-steam power cycle (pressure 30 abs. atm), although preliminary study showed
the economic advantages, in principle, of using combustion heating, which, however, would complicate the arrange-
ment of the station. Five such turbo-units are used, as in the first block; a little modernization enables the power
of each to be raised to 73 MW. Thus the electrical power of the block is 365 MW with an efficiency of 26%.
In order to take off the power from the active zone without reducing the parameters of the steam coming
into the turbine and the water temperature at the entrance to the reactor, it was decided to increase the maximum
enthalpy of the water at the outlet from the most thermally-stressed cassette up to saturation enthalpy, and even
to admit steam content (up to 3 or 4%) in individual hot streams emerging from the active zone. For an average
heating of ? 28?C and water temperature at the entrance into the reactor 250?C, the flow of heat carrier through
the active zone is 42,000 m3/h. This compelled the number of circulation loops of the system to be raised from
six to eight. Abandonment of the idea of repairing separate units in working-reactor conditions considerably sim-
plified the arrangement of the first circuit and made it possible to site the eight loops in two hermetically-sealed
boxes.
In order to maintain the necessary quality of the water in the second circuit, ion-exchange filters are used in
the purification system of the second block of the station, in contrast to the evaporation devices used in the first
block; this enables the gas balance to be stabilized more easily. The filtration is effected while maintaining the
working pressure of the water.
The pressure in the first circuit is maintained by steam volume compensators (not gas as in the first block),
which also aids stabilization of the water-chemical conditions of the first circuit.
As a result of raising the specific and total power of the reactor, a considerable reduction in the fuel and
capital components of the cost of electrical energy was achieved.
The reactor (Fig. 1) constitutes a vertical cylindrical vessel some 19 m high. It consists of a body 10 and
upper take-off block 12 with a plane roof. Inside the body, in a special cylindrical shaft 6, is a removable bucket
7 of diameter 3000 mm and height 4000 mm, serving for the positionining of the hexagonal cassettes containing
the fuel elements 5. The drive mechanisms of the movable cassettes 2 intended to compensate the reactivity and
avert danger in the reactor are set in the casing of the upper take-off block. On a flange of the body is set the
roof of diameter 3350 mm and thickness 520 mm. Into 85 apertures of diameter 100 mm are sealed tubes for the
mechanisms of the reactivity-control systems and tubes for bringing out the cassings of the thermocouples regulat-
ing the water temperature in the cassettes. The roof-body joint is made in the form of a self-sealing lock with a
1184
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
wedge-shaped nickel gasket. Sealing by the flexible element of a
toroidal compensator, one end of which is welded to the roof while
the other presses on the flange of the body, is provided as reserve.
Sealing is here achieved by means of a rod seal of diameter 5 mm
or by welding the toroidal compensator to the flange of the body.
The roof is fixed to the flange of the body by a pressure ring of diam-
eter 4000 mm and height 800 mm with 60 pins.
The outline scheme of the electrical station and the reactor-
cooling system are indicated in Fig. 2. Each of the circulating loops
includes a steam generator, centrifugal pump of the glandless type,
two valves with electrical drive, a reverse valve, and connecting
pipelines of internal diameter 500 mm.
The stitri generators have a horizontal-type design. This
makes it possible to establish the heat-exchange surface below the
flange of the reactor body, so that in recharging the reactor the
circulating loops can be kept full of water, which in turn ensures
the elimination of the residual heat evolution during recharging by
natural circulation, without any special circuit.
In case of a loss of supply in the self-consumption system, a
safety supply system including storage batteries and a Diesel gen-
erator with automatic start is provided in the station. The elec-
trical supply of the circulating pumps of the circuit in the first
seconds after an emergency is effected from the turbine generators
owing to the steam and mechanical coasting of the turbo-assemblies
As the revolutions of the generator fall below the norm, the pumps
are switched on to the storage batteries and subsequently stopped,
the system passing over to shut-down cooling by way of natural cir-
culation in the reactor circuit. The grouping of the turbine and
deaerating sections is analogous to that in organic-fuel electrical-
stations.
Fig. 1. General view of the reactor; 1) bush-
ing for fixing the shaft to the body; 2) con-
trol-system cassette; 3) bottom of shaft;
4) screen; 5) working cassette; 6) shaft;
7) removable bucket; 8) spacing lattice;
9) protecting-tube block; 10) body;
11) temperature-control tubes; 12) take-
off block; 13) drive for the control-system
cassette; 14) servicing platform.
rise
2. WAYS OF IMPROVING WATER-MODERATED
WATER-COOLED REACTORS (WWR)
Improvements to the WWR achieved in the design of the sec-
ond block of the Novo-Voronezh NPS were strictly limited by the
above-mentioned program of modernization of the first block in
this station, and so by no means exhaust ways of improving WWR.
From the point of view of a possible reduction in the cost of
the energy produced, let us consider some other more radical ways
of changing the construction of WWR systems.
Steam-Generation Scheme
It is expedient to retain the simplest arrangement of producing
steam in "boiler" evaporators, supplying saturated steam to the
turbines. The production of power steam by the self-evaporation
of water in the reactor circuit by throttling can only compete with
this scheme at low pressures (clearly not optimum) and with fairly
cheap pumps, the cost of which (per 1 kW pump power), as calcula-
tions show, must be less than half that of 1 m2 heating surface.
Figure 3 shows the dependence of the increase in the relative energy consumption on circulation An and the
in the efficiency of the cycle ATI, after exchanging the steam generator with minimum temperature head .8
1185
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
?
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Fig. 2. Reactor-cooling scheme. 1) Reactor; 2) cir-
culating pump; 3) steam generator; 4) valve; 5) re-
verse valve; 6) air-removing pipeline; 7) discharge
pipeline; 8) pipeline for removing air from the pumps;
9) pump-discharging pipeline; 10) branch to collect-
ing reservoirs; 11) branch to air-purification filters;
12) steam volume compensators; 13) filters; 14) trans-
fer pump; 15) temperature monitor; 16) monitor for
temperature difference in loop; 17) pressure monitor;
18) monitor for pressure drop in active zone; 19) flow
monitor; 20) compensator-level meter.
for a throttle system, on the initial and final pressures. As we see, the increments are only equal (An = Ej, i. e.,
the net efficiency of the system is not reduced) either for steam pressures too low for large powers (p2 = 10-15 abs.
atm) of for unsuitably-large temperature heads Ca > 25?C).
Superheating the steam in the steam generator by Atne leads to an increase An in the efficiency of the cycle
roughly proportional to the product of Atne and the superheating heat qne = CpAtne/Ai. Since the possible super-
heating is proportional to the heating At of the heat carrier in the reactor, the increase in efficiency due to the
introduction of superheating is approximately proportional to At2. Since for WWR small water beatings are best, the
effect of introducing superheating does not, as a rule, justify the complication in the construction of the steam gen-
erator and its greater cost.
Use of a two-pressure cycle raises the efficiency as the first power of it, which makes the increase in effi-
ciency more appreciable even at low At. For example, for At = 40?C, the relative increase in efficiency is roughly
Ai 1 T,At 300.40
? ?0.03,
ii ? ( At
+ ?2)(Ts?T (
x) 40
4 500+ --f-) (500-300)
where Tx = 300?K and Ts = 500?K, the temperatures of the cold and hot sources in the one-pressure cycle.
For uniform heating of the water in the cassettes of the reactor the optimum are heating At rises, as a result
of which the effect may increase ? 1.5 times.
If the difficulties of drying steam at two pressures in the turbine can be successfully overcome, then evidently
for a large power of each reactor-cooling loop it may be advantageous to use two similar steam generators in series
in the first circuit, producing saturated steam at two temperatures differing by 20-25?C. It should be emphasized,
however, that to a certain extent this complicates the arrangement of the first circuit, which should be as simple
and symmetrical as possible, so as to minimize the extent and volume of its pipes and installations and to create
good conditions for the natural circulation of the water in the first circuit and for compensation of the thermal
expansion of the pipes.
1186
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
in,An
C,04,
0,03
0,02
0,01
0
10
0
I?-
4/
at
spat
P1-100
All at
'15
-.4"
h at Pi
7:-.50atill
An at Pi=30
I
Allat
0 =5
10
20
30
plz30
10
20
30
PS
Fig. 3. Increase An in cycle efficiency and An in the re-
lative amount of energy spent on circulation after replac-
ing the steam generator by a throttle system, expressed as
a function of the temperature head in the steam generator
.9. and the pressures in the reactor pi and turbine p2. The
lower graph shows the connection between the minimum
temperature head 9. and the pressures pi and p2 for which
= An, i. e? the net efficiency is unaltered.
Thermal expansion of the pipes may best be allowed for by providing free movement of the equipment, with
rigid, straight pipes. In this case it is convenient to have the pipes of the first circuit, the supports of the reactor,
the pumps, and the steam generator, and the fittings in a single horizontal plane; the main pipes may reasonably
be connected to the reactor above the active zone, which simplifies the construction and improves the reliability
of the reactor body.
Motion of Water in the Reactor
For a downward motion of the water in the reactor, the space above the active zone remains free for carrying
out recharging operations, and the emergency introduction of the absorbing controls into the active zone under the
combined action of the flow and the force of gravity is more reliable. There arises a danger, however, of upsetting
and stagnating the water circulation on an accidental switch-off of the supply to the master pumps and consequent
overheating and damage to the fuel elements, This situation would apparently not have especially serious conse-
quences. It is very hard to check this with the ?necessary degree of reliability. Meanwhile we must confine consi-
deration to a lifting motion (of the water), when there are practically no doubts, but when it is also considerably
more difficult to free the space above the zone and ensure the reliable downwards motion of the controls.
A two-way active zone occupies an intermediate position between oneway zones with upward and downward
motion of the water. In such an active zone, with water being fed in and out underneath, it is convenient to prevent
the cassettes in the central, lifting path from floating upwards by means of hydraulic pistons, and to effect cassette-
by-cassette control of the water temperature only at the bottom?at the outlet from the second peripheral (hot and
downward) path. Then the space above the active zone will be free for recharging without removing the roof.
1187
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
op
10
8
6
5
2
10
grill'
111mini
0 ,.,1.
c4
EllaziNail
q 1
CC A,
11111111M1101.
111?0111111EUIMOMVM
MI
P,
ed-9,
(D c's.
moutramas
''''
ir
IN
__?.,
c 0
z
c c/
0
Maim
Iwo
-mu
iihicc,
- ?1.
..,
6hilli
1
PRI
who
_
IIIIINIIIIIIIIIIII1
10-4
110". 10-3
moopty4
0,10
0,08
0,06'
0,05
0,04
0,03
0,02
0,01
PO4Ta?/
rc
Fig. 4. Variation with auxiliary recharging time ra of the
optimum number of rechargings ''opt per campaign rc and the
corresponding relative increase in the cost of 1 kWh, AC, over
the cost of 1 kWh C? for continuous recharging at full power.
AC(n)/de
0,4
0,2
MIN
I'M N
Ma
Mill MI
t
MI"
111111111=11111
.11111111.11111111.111111.1111.
11111111111?1111111MMINIM
113111111111111111
1111111MEEPPPREIMilifill
? 11.1011M1111.1111111111=11=111
0 10 20 30 qu 50 50 70 80 90 n
Fig. 5. Relative increase in the cost of 1 kWh over the cost
1 kWh with continuous recharging without lowering the power,
plotted as a function of the number of recharging n per cam-
paign Tc and of the cassette-recharging time rr. (Auxiliary
time taken as ra = 10-3 Ta.)
Downward motion of the,water?can be?arranged to take .place in all the.control channels. The two-way active zone,
however, contains the possibility of neutron instability, and as before there are doubts over the reliability of effect-
ing emergency cool-off in the downward path if forced circulation suddenly ceases. The thermotechnical advan-
tages connected with increasing the critical loadings (by a rise in the velocity and underheating of the water at the
danger point) and the possibility of increasing the heating and reducing consumption (by leveling the temperature
of the water in the chamber between passes) are only substantial for a fairly high radial nonuniformity of heat
1188
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
evolution, which in itself is unprofitable. Hence the two-way active zone does not normally have sufficient advan-
tages to justify increasing the hydraulic resistance and complicating the construction.
Ways of Recharging
With high pressure of the heat carrier and a channel-free structure of the active zone, it is most realistic at
the present time to recharge the fuel after removing the reactor roof. In so doing the auxiliary operations occupy
a lot of time (see, for example, [4]), and are accompanied by a substantial change in the "thermomechanical" con-
ditions of the first circuit. Hence it is economically desirable to carry out such recharging as rarely as possible
(not more than a few times in each campaign), despite the fact that more frequent recharging makes it possible to
deepen the burn-up of the fuel being extracted by using the reactivity of that being added [5], and simplifies the
problem of compensation of excess reactivity.
Reducing the duration of recharging lowers underproduction and hence the cost of 1 kWh energy.
Using the approximate connection between the attainable depth of burn-up pe(n) and the number of recharg-
ings per campaign time re (see [5]) Qc (n) Oc (nc?) = I -1- ()-1 , we obtain a relation between the optimum num-
ber of rechargings nopt, the additional shutdown time ra of the WWR in connection with recharging, and the actual
cassette-rechaging time Tr = mrj.:
I ?T'r /To
n=
opt I C? 0 tc ? CF?
a
a , r ta. Cc?
c Or CF
where m is the total number of cassettes, Tj is the recharging time for one cassette,
R R
r dr D3 (r) rdr
o o
Or ? R
C (LI (r) r dr C 412 (r) r dr
(1)
is the
reduction in the depth of burn-up due to the radial nonuniformity of the neutron field .1.(r) [for ci:.(r) = Jo(mr/R),
Or = 1.69], ec and C. are the capital and fuel components of the cost of 1 kWh energy for the ideal condition of
continuous recharging at full power.
For n = nepr, the cost of 1 kWh Ce(n) = Cmin exceeds the ideal cost Ce? = + C? by a fraction
+ Ta or. 2 V___.ta c cg
,Itc ? t 2 1/.0rra/To
AC (flopt) + Cmin?Ce? lc CC?
C? C? ( C? VC?. /C1 + VC? /C? 1+ 114e) F C
These relationships are shown in Fig. 4, while Fig. 5 gives the relation for AC(n)/C, the increase in the cost
of 1 kWh over the "ideal," as a function of the number of rechargings n, from which we see that a reduction of n
to 0.5 nopt does not seriously raise the cost of the power. As we see, the main role is played by the auxiliary time,
reducing which to (10-3-10-4)re may considerably reduce the cost of power. Together with improving the recharg-
ing after removing the roof, the realization of such short times for the auxiliary operations will stimulate the de-
velopment of recharging without removing the roof and shutting down the reactor. '
Recharging the fuel without removing the roof but after stopping the reaction, cooling down, and dropping
the pressure can have little point, since by so doing the only improvement on open-roof recharging would be the
absence of roof resealing, while the recharging itself would be harder and probably take longer. The shutting down
of the reactor before recharging and its subsequent heating have to be carried out quite slowly in order to avoid
thermomechanical damage.
Recharging while running requires complicated-measures?to?restrict the .flow mater through. the cell of the
unloaded cassette and leads to a surge in the neutron field in neighboring cassettes. This form of recharging would
rather demand a reduction in reactor power. Carrying out recharging operations for a critical state of the reactor
also introduces difficulties and seriously retards the recharging process. The use of solid absorbers with upper drive
(2)
1189
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
A-A
8
8-8
Fig. 6. Arrangement of reactor with "honeycomb"
active zone and variation in the concentration of
liquid absorber in the channel system: 1) body;
2) liquid-absorber collectors; 3) tubes for cassette
control of the hermetic state of the fuel elements
and water temperature; 4) driving-mechanism rod;
5) roof; 6) recharging machine; '7) diving system;
8) unit securing the cassette from floating up.
1190
is entirely excluded; use of lower drive is limited by incon-
veniences associated with the disposition of the absorbers
above the active zone. In view of this, choice of a reactivi-
ty-compensating system evidently reduces to the use of liquid
compensators, i. e., changing the concentration of an ab-
sorber in solution. For this the most appropriate is probably
an active zone developed into cells (kind of honeycomb) by
special barriers, which, however, occupy a perceptible part
of the useful volume and constitute quite a skeleton struc-
ture (Fig. 6).
Without removing the roof it is most suitable to re-
charge the fuel without lowering the temperature and pres-
sure, but after switching off the reaction. This kind of re-
charging can be effected in two stages: 1) unloading from
the active zone into intermediate storage cells (or inversely);
2) removing the cassettes from these cells through a gate in
the body (or loading new cassettes into them). The more
complex second stage may be effected while the reactor is
running without time limitation.
The structural arrangement of this recharging is anal-
ogous to that shown in Fig. 6. Inside the body of the reactor
are movable or immovable intermediate-storage cells for
cassettes outside the active zone, These cells and the gate
in the roof lie one over the other, thus easing the kinematics
of the external charging machine. The container of this
may be furnished with a "magazine" so that the gating only
has to be done once per recharging.
For n rechargings per campaign, the number of inter-
..mediate-storage cells must be roughly 1/n-th of the num-
ber of cassettes in the active zone (for 300 cassettes and 20-
100 rechargings, there should be 16 or 4 intermediate cells
respectively).
If the exchange time for one cassette is r1 = 15-30min,
then for 300 cassettes and a campaign time Tc = 500 days the
increase in the cost of 1 kWh of energy owing to deviation
from continuous recharging at full power is an extremely
small quantity:
AC (nopt)
1
nrril
1
300 (0,25 ? 0,5)
? 2
tc ?
2
500.24
(0,3 ? 0,6) %.
Estimating the length of the auxiliary operations
(switching off the reaction, freeing the space above the
active zone from the drives of the control systems, and
restoring the working state) as Ta = 5 h, for er = 1,69 and
ec/C?F = 1 we obtain:
C?
fl opt 170r F = y1,69.500.245 1,65,
? T C?
a C
and the cost of 1 kWh further increases by
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
AC (nopt) 2 liertatrc = V1,69 0
co ? Co /- 500.24_ 2,6 96.
V_2,6
Cc? C?F
On placing the drives underneath the active zone, the auxiliary-operation time Ta may be reduced, and this
may diminish this figure.
Fixing the Cassette
In a spiked working reactor, the lifting force of the current flowing from bottom to top may several times
exceed the weight of the cassettes, and these must therefore be braced against floating up or relieved from the pres-
sure drop. In the presence of some kind of spacing lattice between the cassettes, these may be fixed by spring catches
to the lattice walls (position 8 in Fig. 6), or by using catches or spring clamps in the tail of the cassette for fixing in
the lower supporting lattice. These methods required either an increase in the space between the cassettes, or greater
force in the recharging machine, or lock-control system passing through the active zone.
A more rational method of relieving the cassettes is a system of pistons fixed to the lower part of the cassette
and experiencing an inverse pressure deop, due to the fact that a flow-collecting chamber, connected by a bypass
tube to the space above the active zone, is formed under the pistons.
All the methods described make it possible to free the upper part of the body from apparatus preventing the
recharging of the fuel.
Control and Protection Systems and Drives
Solid absorbers placed along the axis of the active zone are most widespread in WWR. In both the reactors
of the Novo-Voronezh NPS, control cassettes with absorbers furnished with neutron water traps are used.
Replacement of the compensating and emergency absorbers by fuel when raised ensures a good neutron balance
and increases the compensating power of the control elements, but on the other hand causes difficulties in the rapid
introduction of the absorber against the direction of motion of the heat carrier. Hence it is very necessary to relieve
the control cassettes from the action of the upward-moving current in the present case for high specific heat extrac-
tions. Such control elements introduce considerable nonuniformity into the distribution of the neutron field for inter-
mediate positions, and therefore, generally speaking, it is appropriate to use them for emergency protection and the
compensation of changes in reactivity due to the temperature effect, the Doppler effect, and poisoning.
It appears more justified to leave compensation for fuel burn-up to absorbers of relatively small efficiency,
the movement of which does not introduce substantial nonuniformities into the neutron flux. It is possible, for ex-
ample, to use lamellar and cylindrical absorbers furnished with scattering tails. For a fairly high recharging fre-
quency (meaning recharging without removing the roof), the excess reactivity compensating the burn-up may be
small, so that the over-all efficiency of the absorber need not be high.
We cannot exclude the development of combined absorbers, combining the principles of absorption in traps
and in thin plates. The use of liquid absorbers, the concentration of which in the moderator or special water circuit
may be varied, also deserves attention. In WWR conditions the compensation of reactivity by charging the neutron
spectrum may prove hard to realize.
It is appropriate to use the following drives for solid control- and-protection systems.
1. Electromechanical or electromagnetic drives. These are placed in sheathes welded into the roof and con-
nected with the working part by a rigid rod through a remote-release coupling. The sheathe is tight, i. e., the en-
ergy is fed to the drive electromagnetically through a close nonmagnetic screen.
During recharging, the drive rod is disconnected from the working part and taken away up, while the working
part remains in the lowest position. The absorber projecting above the active zone and the rigid connecting rod
may require protection from the action of the flow of water coming out of the active zone.
This scheme of electrical drive is similar to existing ones and is accessible for inspection through the upper
flange of the sheathes in the roof, but in principle it is not suitable for recharging while running; it dces not ensure
emergency protection for recharging without roof removal, occupies a large space in the roof, and requires a re-
serve of travel upwards for removing the rod from the recharging space.
1191
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Fig. 1. Hydraulic drive
system: 1) supporting
plinth; ) plinth With
hydraulic-cylinder
Sheathes; 8) cooling'
water entrance; 4) pis-
ton-tail Working part
(upper position); 5, 6)
position indicators;
1) damper tube; )water
supply under the piston;
9) to position indicators;
10) piston cibWri.
1192
2. Hydraulic drive situated beneath the active zone (Fig. `7). A piston is con-
nected with the "tail" of the working part and moves in a hydraulic cylinder under
the control of water pressure under the piston. There are units for detachably con-
necting the piston to the tail of the cassette and the cylinder to the controlling water,
units for limiting the extreme positions and fixing intermediate positions, and also a
unit for reStricting the rate of lift. This kind of drive allows rapid clearing of the
space for recharging without disconnecting from the working parts, and retaining the
emergency protection during the recharging period:
The hydraulic drive demands extremely reliable coupling of the piston to the
fuel working part, so as to eliminate the possibility of its floating up under the action
of the flow. This fact forces us to consider combined drive's with a mechanically
movable support above, fixing the intermediate positions. This system is quite cum-
bersome, but facilitates a rapid downward throw of the absorber from any intermediate
position. One may connect a hydraulic drive to the nonfuel elements of the control-
and-protection system, 'which are not threatened with floating up under the action of
the cooling flow of water, or go as far as a nondetachable joint of the Working part
With the hydraulic piston.
Compensation of Burn-up by Lowering the.PoWer
There is some intereSt in the possibility of raising the burriiip and lowering the
cost per kWh by gradually reducing the power of the reactor, maintaining the param-
eters, so as to compensate burn,up by reducing the DOppler effect (and partly dimini-
shing poisoning). This decreases the fuel but increases the capital Component, since
less power Will be produced in the profit period, Hence there ekiStt an Optimum
lowering of power, as calculated below,
In the linear approximation, the cost of 1 kWh may be expressed in the form
Ce(tic)--=CF+ Cc = + k (1 ? 3.01-1+ ? nk ln RI [1 + nk (1 ?
(3)
where Co5 and ec ar the fuel and capital components of the cost of 1 kWh for opera=
AlC(42
tion at nominal poWer and depth of burn-up pet); k =--- is the re-
Qt Q( (aK/a@c)
lative increase in burn-up resulting from the use of reactivity Aleo released on re-
ducing the power to zero; n is the number of rechagings in campaign time tc for
power Q0; x = = exp(--rakk) gives the extent to which the power is reduced
for the additional working time ex The components of the Colt of 1 kWh May be
expressed in tertris of CF (dolt of 1 ItG Of fuel) and k (capital expenditure in 'setting
up the NPS):
This cost (.yt.) for n = 1 is smaller than Ce( x = 1) = = + t by a file-
tion (AC/Ce?) = kX[1. + (1 + CVC?0-11ril.d[i + k(1 14.)]'1; which in the eaSe
of k(1 ?R) 1 is it-laid-mum for - opt = (1 + p/C)=1: For exarriple, for CVC!L
= 1 we obtain xopt = 0.5, Te 5t = toklnv,-1 = 0.6910-c and (6,Ciee)max = 0,16ki(1+0,15k).
Hence for k = 0,4 the advantage in the cbSt of 1 kWh is 510, and the duration of
operation at reduced power is = 0.28 rc.
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
For a large number of rechargings the advantage in the cost of 1 kWh will be smaller. In the limit n-0. *3,
instead of (3) we obtain for the cost of 1 kWh Ce() = C[l + k(1 ? x)]-1+(C? lnx*1)/(1 ? x), which will be lower
than C?e = C + ec only for fairly large values of k or C/
C. C?c. For example, for C?F/C?c = 1, the value of k=
should be greater than 0.5.
Although the improvement in cost is not very great, we must remember that this is a gain in comparison with
the cost corresponding to continuous recharging, since the unproductive use of neutrons in compensation by the ab-
sorbers is excluded. Furthermore, use of the power effect simplifies the control system thanks to the reduction in the
number of burn-up-compensation controls, and makes it possible to increase the time between rechargings, i. e.,
to raise the chances of charging systems with roof removal.
3. OPTIMUM WWR PARAMETERS
If particular types of main components are selected for the NPS and the reactor, then optimization of the
parameters characterizing these units quantitatively may also substantially lower the cost per kWh.
Since the number k' of independent parameters ne is large, and their connection with the cost of energy com-
plicated, direct optimization by simultaneous solution of a closed system of equations (minimum cost of 1 kWh cor-
responding to oce/axk, = 0) is only possible in exceptional cases [6]. The form of these conditions is often complex,
since in varying each parameter all the rest in the chosen set are kept constant, and not just those "convenient" com-
binations in fixing which the form of the conditions ace/oxit, = 0 becomes simple. If we select such combinations,
then we may find simple relations between the optimum parameters, reducing the voluming of computing work by
many orders. In particular, in a number of cases it is convenient to fix the power of the reactor and, together with
this, part of the electrothermomechanical equipment of the NPS.
A considerable simplification will also be achieved if we consider that normally the relative changes in the
cost per kWh Ce are much smaller than the relative changes in the independent parameters being varied. Such pro-
cedures were developed in [7] and applied to WWR of the type considered.
All the parameters were divided into three groups. The parameters of the first and second groups at the opt
mum are expressed analytically in terms of those of the third original group. In this are included the cost coeffi-
cients of the equipment, and also the volume fraction of water in the active zone e, the pressure pi, the coefficients
of nonuniformity of heat evolution, heating, and velocity, and other parameters the optimum value of which has not
yet been obtained in analytical form.
In the first ("thermomechanical") group come the power Q, the zone length 1, the heating of the water in the
active zone At, the head .9. in the steam generator, and other associated parameters. These quantities are optimized
for a fixed cost of fuel, determined by the parameters of the second group.
The second ("fuel") group comprises the diameter dF, the depth of burn-up pc, and the length of the campaign
Tc for the fuel elements. These are connected by the definition pc =rcpsp(dF) in terms of the specific power psp
4(1F/YLI3F 4c1//YUIrdF?
The heat flow from unit length q1 or surface qF of the fuel element must, at the "stressed point," be kept
a
equal to the limiting permissible value qa i or q i
F ; f q > q3, then some advantage may be gained by making these
flows equal to each other, increasing, for example, the corresponding diameter of the fuel element, which
lowers the cost of their manufacture and the amount of covering for the same power and otherwise equal conditions.
For fixed parameters of the first group (including the power and the parameters of the heat carrier and the
steam), the capital expenditure and efficiency will be constant, and the minimum cost per kWh will be obtained
for those values of pc, rc, and dF which make the fuel component a minimum.
Analysis of the relations obtained in this way in [7] for a cycle without repeated use of fuel leads to the fol-
lowing:
1. In the region of fuel-element diameters* dF do = q2/71-4 in which the linear heat flow equals the limit-
ing value q1 = q, the optimum burn-up popt is determined only by the relationship of CF m(pc), the cost of the
a a
*Here do = hrq i
F s the diameter for which the heat flows from unit length and unit surface of the fuel element
a a
simultaneously take their limiting values, q1 = qz and qF = qF .
1193
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
fuel material, to the depth of burn-up, and lies at the point pc = P? Pt'
pt, in which the tangent to the curve CF m(pc)
passes through the origin of coordinates (1. e., 3C C Here the optimum diameter is proportional
a
to the complex [qi kmanuf/CF m(pc)pc]-71/4, increasing i\cre4a/Psing depth of burn-up and cost of fuel material,
and with decreasing cost of manufacturing unit length of fuel element (0.257ry ukmanuf, where yu is the specific
gravity of the fuel).
This creates a tendency toward the transition of the optimum dF into the region dF < do.
qpa
2. In the region dF do the heat flow qF equals the attainable , and the optimum burn-up is somewhat
smaller than the quantity pn'pt, tending toward it with increasing 4 and decreasing cost of making one meter of
iacF
fuel element as 1/7?/
Vkmanuf. Here the optimum diameter varies as
11 aQc
kman ?Qc acF'm
a
uf isc Qc ? ?
3. The optimum values of pc and dF are sensitive to the relation between the cost CF m(pc) of the fuel being
loaded and the depth of burn-up of the fuel being extracted. Apparently for dioxide fuel the optimum in a number
of cases lies in the range 30-50 kg/ton, and the optimum diameter is not very different from do = 4/7r 10 mm.
r
The optimum campaign is more stable. It lasts around 103 days and increases as the complex (kmanuf)Y2/[q?
(CF m/pc)1V2 with increasing cost of manufacturing fuel elements and decreasing values of the attainable flows and
ratio (CF m/pc).
With the aid of the corresponding relations of [7], 288 sets of optimum parameters of the first group were cal-
culated for different original parameters of the third group, which contain certain arbitrary features and were there-
fore varied over wide limits.
In this the following was taken into consideration:
1. The maximum heat content on emergence of the water from the cassette must not differ greatly from
saturation imax (p = Imax < then the enthalpy at the inlet iin has to be lowered or the flow rate in-
creased (in both cases efficiency falls), or the power reduced. These negative effects are not compensated by a
certain increase in the critical loadings on increasing the underheating up to boiling in the danger region above the
middle of the active zone.
For imax > i' there appears a steam content x =:
?max ? which worsens the neutron balance and reduces
the hydraulic stability of the parallel operation of the channels; hydraulic shocks arise on condensation of the steam
after mixing. For small x these effects are apparently not very great, but the quantitative aspect of the problem is
not quite clear,
2. The diameter of the active zone should be made a maximum (? 3 m), allowing for its being placed in a
works-manufactured framework of maximum transportable dimensions (Dtr 4 m).
Increasing the diameter above Dtr would be advantageous if it were not accompanied by a reduction in the
thickness and strength of the body material and a corresponding sharp fall in the pressure and efficiency, which are
not compensated by a rise in power (in boiling-water reactors the situation may be different).
Reducing the diameter leads to a rapid fall in power, by no means compensated by the reduction in the cost
of the body and the increase in efficiency for a rise in the permissible pressure, which even for D = 4 m is close to
optimum. As the pressure rises, the efficiency increases all the more slowly and the critical loadings fall all the,
more rapidly, which makes it necessary to lower the power or increase the surface of the fuel elements. For
ackrit/aPi= ?0.03.106 kcal/m2? h ? abs. atm [8], the optimum pressure is ? 120 abs ? atm, which is still realizable
in a transportable body. Achieving critical and hypercritical parameters, which may sharply improve many pro-
perties, is another matter. This is not considered here.
The results of applying the relations obtained reduce mainly to the following, valid at the point of minimum
cost per kWh (these admit, as a rule, a fairly clear physical interpretation, although sometimes "common sense"
could promp other, false tendencies):
1194
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
1. The effect of the water-uranium ratio Nu on the capital component of the cost per kWh in the optimum
proves extremely weak, despite the strong variation of the fuel-element surface and the active cross section of the
heat carrier with changing Nu; this is due to the compensation of the Nu effect by the inverse effect of the relative
length of the active zone //dF, which at the optimum rapidly increases with rising Nu [roughly as Ncti?}6(1 + Noot.
2. The optimum thermal power Qopt depends mainly on the radial coefficient of nonuniformity of heat evd1u-
tion Kg, constituting 5-6 million kW (thermal) of Kg = 1.5, and3.3-4 x 10kW for Kg = 3. ApproximatelyQopt ?Ka
and is practically independent of the admissible heat flows, an increase in which may more suitably be used for re-
ducing the relative length of the active zone.
3. The average heating of the heat carrier at the optimum depends most of all on the nonuniformity of heating
Kt =Atmax/Atav (roughly topt KA-3/4) and is 45-65?C for Kt = 1.2 and 30-45?C for KAt = 2.
4. The optimum average velocity of the water in the active zone Wopt is mainly determined by the nonuni-
formity of velocity Kw =Wmax/Wav? and is 3.5-4 m/sec for Kw = 2.5, and 6-7 m/sec for Kw = 1.25 (approximately
opt W ?
The maximum heating values Atmax and especially velocities Wmax depend on Kw and Kt only weakly:
= K AtAt opt-- KS:t2
max 5
At
and
Wmax =-KwWopt
Wmax changing in general very little (within 8-9 m/sec).
On varying the density of heat take-off p from unit volume of heat carrier, the velocity and heating Of the
water may advantageously be changed in the same way as Arp.
5. The optimum fraction of energy spent on pumping over the heat carrier rip = Np/Q depends fairly weakly
on all the original varying data and forms 0.6-0.9% of the thermal power.
6. Also stable is the optimum average temperature head in the steam generator, equal to ? 25?C for a non-
corrosive heating surface.
7. Finally, the cost of an established 1 kWh, Cest, (without the cost of the first charging) is determined to
a considerable extent by the relation of the cost of the second circuit and auxiliary systems of the NPS to power;
cNPS = A + BQ. The chief part is played by the coefficient B, but it is interesting to note that this has a very weak
effect on the thermomechanical parameters being optimized. The value of Gest falls considerably on reducing the
coefficients of nonuniformity K and Kt.
With increasing pressure, the value of Gest also falls, but the length of the active zone and the charge of
fuel increase (owing to the fall in the permissible heat flows). Hence there is an optimum pressure pi of the order
of 120 abs? atm, as was discussed earlier. For this pressure and a maximum heat content of the water corresponding
to saturation, an average heat flow at the surface of the most-stressed fuel element qmax = 1.2- 106 kcal/m2?h
? oC,
and a radial heat-evolution nonuniformity Kg = 1.5, the optimum parameters of a WWR in the maximum transport-
able framework of diameter 4 m lie roughly within the following ranges: thermal power Q 5.5- 106 kW/m2; ratio
of the length of the active zone to the diameter of the fuel element ? 500 (for a water-uranium ratio of ? 1.6);
mean heating and velocity of the water 48?C and 6.5 m/sec (for Kt = 1.2 and Kw = 1.25); mean head in steam
generator 26?C; steam pressure p2 45 abs ? atm; efficiency of cycle ? 30%; proportion of power spent on circula-
tion nc 0.65% (for pump efficiency ? 0.7).
The parameters of the second block of the Novo-Voronezh NPS are close to the optimum values correspond-
ing to the pressure, nonuniformity coefficients, and power of the reactor in this block (Q stzs 1400 MW), which, how-
ever, is considerably below optimum.
LITERATURE CITED
1. Planning of 420 MW Atomic Power Station. Report to the 11th Section Meeting of the All-World Power Con-
ference [in Russian], Belgrade (1957).
1195
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
2. S. A. Skvortsov, In the book "Transactions of the Second International Conference on the Peaceful Use of
Atomic Energy," Contributions of Soviet Scientists, Vol. 2 [in Russian], Moscow, Atomizdat (1959), P. 105.
3. R. S. Ambartsumyan et al., Ibid;, p. 119.
4, P. Eddy, Nucleonics, 21, 101 (1963).
5. S. M. Feinberg et al., See [2], p. 411.
6. A. Ya. Kramerov, "Atomnaya energiya," 10, 211 (1961).
7. A. Ya. Krarnerov and Ya. V. Shevelev, Engineering Calculation of Nuclear Reactors [in Russian], Moscow,
Atomizdat (1964).
8. J. Aladyev et al., Boiling Crisis in Tubes, International Developments in Heat Transfer, Vol. 2, New York
(1961), p. 237.
1196
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
THE ORGANIC-COOLED ORGANIC-MODERATED
NUCLEAR POWER STATION "ARBUS"
K. K. Polushkin, I. Ya. Emel'yanov, P. A. Delens,
N. V. Zvonov, Yu. I. Aleksenko, I. I.,Grozdov,
S. P. Kuznetsov, A. P. Sirotkin, Yu. I. Tokarev,
K. P. Lavrovskii, A. M. Brodskii, A. R. Belov,
E. V. Borisyuk, V. M. Gryazev, V. D. Tetyukov,
D. N. Popov, Yu. I. Koryakin, A. G. Filippov,
K. V. Petrochuk, V. D. Khoro'shavin, N. P. Savinov,
M. N. Meshcheryakov, V. P. Pushkarev, V. A. Suroegin,
P. A. Gavrilov, L. N. Podlazov, and I. N. Pogozhkin
Translated from Atomnaya Energiya, Vol. 17, No. 6,
pp. 439-448, December, 1964
The idea of utilizing low-rating nuclear electric power generating stations was a corollary of the need to sup-
ply electric power to remote and barely accessible areas of the USSR where the building of conventional electric
power generating stations would not be justified economically principally because of the high cost of fuel delivery
or the high cost of mining fossil fuels in the area itself. Engineering costs calculations show that small scale nuclear
power utilities could be economically competitive in such areas even today. As is generally known, the cost of
electric power produced at a nuclear power station is characterized by a relatively high capital investment com-
ponent, and this holds particularly for low-power electric generating stations. The capital investment component
can be lowered to manageable proportions through the use of organic coolants, by relying on cheaper structural ma-
terials, through the use of mass-produced ancillary equipment and instruments, and by weight reduction or even com-
plete elimination of the biological shielding around the primary loop. To date, however, the widespread use of
organic coolants in power reactor design practice has been held in check by several undesirable effects associated
with radiolytic processes which affect organic compounds. Prominent among these effects is the buildup of high-
boiling- point products of radiation-induced polymerization in the coolant, a situation which could eventually lead
to the formation of insoluble compounds precipitating out in film form on the heat transfer surfaces, with a resulting
vitiation of the heat transfer performance of the coolant. As experience in the operation of the OMRE experimental
reactor in the USA has demonstrated, the simplest cleanup of the coolant by distillation and rectificatiun to get rid
of the high-boiling polymerizates will ensure that the concentration of these substances is kept at or below a speci-
fied level, but still will fail to prevent deposits from settling out on the surface of the fuel elements. Moreover,
this approach to coolant cleanup implies the need for continuous makeup of fresh coolant supplied to the loop arid
the disposal of high-boiling insolubles extracted from the loop. This imposes a troublesome limiting factor on the
choice of suitable organic fluids, given the heightened requirements of radiation stability.
The brunt of the efforts spent in dealing with the problem of how to best use organic coolants in nuclear power
stations was therefore directed to the search for ways to recover the high-boiling radiolyzates without first extracting
them from the loop, which would open the door for the use of inexpensive and accessible products exhibiting a fairly
low radiation and thermal stability as reactor coolants. These efforts culminated in the design of a regeneration
scheme based on the catalytic hydrocracking process. Preliminary test-loop studies indicated that this process suc-
cessfully hydrates unsaturated products of radiation-induced dehydrogenation and effects a selective degradation of
high-boiling-point radiolysis products, given a judicious choice of process parameters. This method of loop cleanup
rendered possible the use of hydrogenation-stabilized gas oil produced on the basis of a directly distilled kerosene-
gas oil cut of naphthene-aromatic base petroleum as organic coolant for the first nuclear power station. Added to
*Paper No. 307 presented by the USSR delegation to the Third International Conference on the Peaceful Uses of
Atomic Energy, Geneva 1964.
1197
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
the generally familiar advantages inherent in Otganic coolates coolants are such advantages of gas oil as the low
freezing range (-40?C to ?70?C), freedom froth any required use of a loop heating system, and the low cost.
Below appear some salient characteristics of gas oil:
Specific Weight at 20?C; g/ctn3
10dine number
0.8558
Not greater than 1
Total sUlfonating components; w/
30
Low point on boiling range, ?C
212
High point on boiling range, ?C
300
Carbon Content,
86.89
Hydrogen cotitein,
13.11
I-1/C ratio
1.8
Sodium content, w/b
.
210-5
Sulfur content, 10
3.10-3
Vapor pressure at 350?C3 atm
4.85
Chemical tompoSitiOn:
paraffin hydrocarbons, 10
30:12
aromatic hydrocarbons,
tiaphthene hydrOcarbOn?, 10 ?
3003
39.85
The first Af213US type hitcleat.power generating Statibh Was built at the Melekets Atomit Reabtot Reseatch
inStitute: The ptirit ipal Station paratileteis are the following:
Reabtdi poWer output, MW ?
5000
ttirbbgeherator oinplit, kW
150
Pressure in priniary- loop pres?tirizer; atrii
Cdolarit temPerathre at reactor entrance, 30
Coolant tertiperattire at teactor exit; ?C 243
Codlant flOWiate iti ptitriaty lbop, tOns/h 600
Saturated steam teniperaitite in steam generator, ?C 223
The physibai ?taittip of the Organic-beetled reactor in this power staticih tdok place On Jude 293 19633 and the
station went on the line after heat transfer testi had been ebriipleted oh August 11, 1963.
Basic RoWer Statibn Layout.
,.. ? ?.
The pOWei station Operates on two 166ips (Fig. 1): The criolant is Circulated through the primary loop by two
electrically dtiven pumps WOrking oh tWO parallel flow branches Which are reuttited by the reactor. Each pump has
a throughput Of 430 Cubic Meters per haul' at a head of 43 Meters liquid colunith The electric motor develops 50 kW
power. The coolant flows but of the reactor ihto Steam generator units featuring a free evaporation level: the ad-
Varitages of these steam generators are their Simplicity, reliability, atid less stringent feedWa.ter reqUirerriehts. After
fejedting heat to the secondary- loop Water in the Steam generators, the Coolant then proceed? through the pressut-
izeis Which Sittiultanebusly fulfill the fundtibia of degasifiets: Degasificatidn takes place by virtue of the free level
of the coolatit add the special degasifyitig columhs to Which 10106f the total Coolant flow is deiiVered. Mesh fillets
tot rough purifiCation ate also installed in the pressurizer; Froth the pressurizer, the eoblant is taken up by the cif
culating Putrips arid recycled to the teacter. The primary-loop pressure is kept constatit during plant operation by
the gases liberated in radiolySis of the coblaht; hitrogeri is Used to Maintain the pressure when the power statioh is
Started up Cold: Excess gas is Vented to the atmosphere by a pressure tonti011er. Removal of residual heat release
in the reactor hi case Of outage Of the primary- loop tireulating pumps-during the initial period is taken care of by.
two turbine purnOS Widtkitig on the steani acciimulated in the steani geherators; this steam Can keep the turbine
pumps going tot 00 riiiri. The cbtilarit fibWrate is 96 tonsth in that case. Further heat fembval is handled by the
hattirai circitiation of the cablaht:
The pritnaryLlbop coblatit is purified by means of cermet filters installed on bypass circulating pumps. The
filters retain suSpehded pattithilates greater than 1.5 to 3j across and preVeht the concetitratien Of iron in the coolant
freirn SurpaSsing 0.3 mgiliter. The flow thrcitigh the cern-jet filters comprises 1010 of the total coolant flog.
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
?
A
??,\
\1 12
23
4
33
32
30
23
14
22 21 20 19 18 1 71 16
23
Fig. 1. Basic flowsheet of the ARBUS power generating facility: 1) steam generator; 2) basic primary-loop piping;
3) pressurizer; 4) emergency turbine pump; 5) main circulating pump; 6) gas piping; 7) reactors; 8) removal of light
gas oil cuts; 9) receiver; 10) control valve; 11) nitrogen cylinder; 12) water in, from chemical water treatment plant;
13) steam lines; 14) pressure relief and cooling unit; 15) turbine with reducing valve and generator; 16) circulating
pump; 17) condenser; 18) hotwell pump; 19) deaerator; 20) electric feed pump; 21) turbine feed pump; 22) second-
ary-loop piping; 23) cermet filter; 24) metering pump; 25) regenerative heat exchanger; 26) circulating compressor;
27) electric furnace; 28) catalyst reactor; 29) cooler; 30) mesh filter; 31) gas separator; 32) booster compressor; 33)
hydrogen lines; 34) electrolyzer; 35) makeup pump; 36) auxiliary gas oil lines; 37) drainage tank; 38) overflow tank.
The primary loop and makeup for the primary loop are rounded out by a pump operating from an overflow
tank of 20 m3 capacity. Coolant overflow will be cycled to the overflow tank or to the drainage tank depending
on the degree of contamination. Low-boiling gas oil fractions forming during the process of decomposition of the
coolant (these fractions exhibit boiling points to 120?C) become condensed in the receiver, whence they are dis-
charged periodically into the drainage tank. Coolant is bled from the primary loop for regeneration, and the re-
generated gas oil is cycled to the overflow tank. A description of the regeneration system and the operating prin-
ciples of that system are given below.
The secondary loop comprises part of a conventional condensation steam turbine electric power station. In
the event of a sudden drop in the turbogenerator load, a reduction cooler unit is provided for the purpose of cycling
excess steam directly to the water-cooled condenser (Fig. 2).
Power Station Equipment
ARBUS is set up in discrete completely modularized units which had been run through test-stand checkouts at
the site of fabrication. The power station is made up of 19 units none of which weighs more than 20 tons. The to-
tal weight of the overall plant, biological shields around the reactor included, is roughly 360 tons. The weight and
1199
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Fig. 2. Steam power equipment of the nuclear station. Turbogenerator appears up front.
sizes of the modular units facilitate transportation of the units to the construction site by water or overland routes.
Assembly of the power station on the construction site was completed in a matter of two to three months,
ARBUS is housed in a building covering 12.36 by 28.5 In, and standing 6.36 m high. The electrolyzer and the
overflow tanks are located outdoors. Figure 3 presents a clear picture of the layout of the equipment inside the
building.
When the power station is being started up, electrical equipment is supplied with power from a diesel gener-
ator of 135 kW rating, so that the system operates in a fully autonomous manner. Natural circulation flow is used
in heating up the primary loop and emergency cool -down of the primary loop, because of the different levels on
which the reactor and the steam generator are located.
The power station equipment includes a fuel element container, a guiding mechanism for seating in7 pile
components, specialized 'hardware for reactor refueling operations, a storage rack for spent fuel assemblies and con-
trol and shielding sleeves with rods. The reactor is refueled with the aid of a special bridge crane boasting a load
carrying capacity of 12 tons.
The reactor and plant staff numbers 17.
The equipment, valves and fittings, and piping of the primary loop are made of low-carbon steel. Mass,pro-
duced petroleum industry pumps and standard petrochemical fittings with tightened specifications on the finish of
the interior surfaces Are employed in the plant. The nonstandardized equipment is made of St-20 structural steel
with modularized shells and bottoms for the reactor, steam generators, and pressurizers. The piping was welded in
a protective argon?carbon dioxide environment.
Maximum ,estimates of the activity of the primary-loop coolant showed it to be 1.5 -10'4 curie/liter even
taking -into account a possible escape ,of fission products from ;ruptured fuel elements, as Idetermined by deliberate
1200
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Fig. 3. Overall view of ARBUS power generating plant:
1) reactor module; 2) steam generator module; 3) tur-
bine module; 4) condenser module; 5) deaerator module;
6) regeneration module; 7) high-pressure gas module;
8) auxiliary pumps module; 9) spare water tank; 10) con-
trol panel and data display; 11) electrical gear; 12) con-
trol rods panel; 13) storage rack for spent fuel assemblies;
14) main switchgear; 15) diesel generator; 16) diesel unit
control panel; 17) chemical water treatment module.
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
puncturing of the cladding in loop tests. This made it possible to do without biological shields around the primary
loop, and to retain only the reactor shielding made of conventional shielding materials (concrete, graphite, poly-
ethylene, iron).
When the makeup of the central power station equipment and the operating conditions of the equipment are
taken into account, the radiation level in rooms occupied by personnel were estimated at 0.51jrem/sec, and that in
the piping system at 3 to 41irem/sec. The activity of the gas effluents formed as a result of racliolytic decomposi-
tion of the coolant, was 10-12 curie/liter.
The total activity of the gas vented by the power station was approximately 7.2 ? 10-8 curie/day.
The Reactor
The reactor is a welded cylindrical vessel standing 4365 mm high, extending 1340 mm in diameter, with a
wall thickness of 20 mm, a flange with support lugs and eight pipe connections 150 mm in diameter for inflow (four
bottom pipes) and outflow (four top pipes) of coolant. Side and bottom shields were installed to lower the irradiation
intensity of the vessel.
An internal vessel is placed inside the reactor to organize the flow of coolant, and at the same time to support
the weight of the core. The uniformity of flow distribution ahead of the core is achieved by means of two perforated
baffle plates.
The fuel elements are clad with low-copper D-20 aluminum alloy. Uranium-aluminum alloy used for fuel
minimizes escape of fission products into the coolant loop in the event of fuel element rupture. The total
U288 loading in the reactor is 22.5 kg, with 36% enrichment. The maximum fuel element core temperature is 336?C
by rating, that of the cladding 335?C.
The physical design of the ARBUS reactor core was based on the results of earlier criticality experiments per-
formed on experimental channels of a VVR-M reactor, a description of which may be found in the proceedings of
the second Geneva conference on the peaceful uses of atomic energy. The choice of these channels is dictated by
the similarity to the channels of the ARBUS facility both in geometry and in the specific U235 loading per running
centimeter of the channel length.
During the physical startup of the ARBUS reactor, the critical level of the moderator was determined for the
fully assembled core, and determinations were made of the number of control rods required to override the total
excess reactivity, the neutron flux distribution, was mapped, and rods were calibrated in a hot clean system. The
results of the physical startup showed that the procedure relied upon for calculations provided the accuracy required
in determining the physical characteristics of the core. Adequate agreement between theoretically predicted and
experimental data was attained in the determination of the critical height Hcr required for a "clean" core at room
temperature (HP= 314 mm; Htheorer =320 mm) and the position of the control rods in critical cold clean and hot
clean systems.
Experimentally measured neutron fields, measured over the radius and height of the core, confirmed the theo-
retically computed values of the variation factors used in the heat transfer calculations.
The system was heated up from 20? to 200?C using an external heat source, in the determination of the tem-
perature effect and of the temperature coefficient of reactivity. Extrapolation of the results arrived at to the oper-
ating temperature of the moderator (245?C) revealed that the experimental value of the temperature effect from
20? to 245?C exceeds the theoretically computed value of 6.1 ? 10-2 by approximately keff= 1.4 .10-2. The temper-
ature coefficient of reactivity is negative over the entire range of temperatures concerned, and is ?4.3 ? 10-4 11?C
at the operating point.
Taking into account the corrections obtained in the physical startup and in the power startup, the values of
the effective multiplication factor for the various states of the reactor are the following; at the beginning of the
reactor period, keff =1.264 for t =20?C; in the case of a hot clean reactor kat- =1.189; in the case of a hot poisoned
reactor keff = 1.148. The duration of the reactor period with the reactor on full power is about two years.
Reactor control is achieved by means of cylindrical rods traveling through the reactor core. Two boron steel
rods are designed for automatic control (the automatic control rods); 30 boron steel rods are designed to override
the temperature effect and the poisoning effect (reactivity compensation rods). These rods compensate ?12% of
the reactivity.
1202
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
26,5 2800r250
26,0 2 680
15,5 4_71560
8
a.
o
(2.: 150 8244
a.)
1.)
a. rx r.4
03 24,5 2320
24,'220
gZ..1
'4.00
. 1
/
I
4
17 5
I
o
I
11
o
o
I
150
I
I
725
I
1
o
o
300
140
130
110
110
100
0 300 600 900 1100 t, sec
Fig. 4. Variation in power station parameters in response to increased
power load: 1) electric power; 2) reactor power output; 3) steam pres-
sure; 4) coolant temperature at reactor exit.
Unc'er emergency conditions of the first kind (scramming conditions), (i.e., cessation of coolant flow in the
primary lcop or reactor runaway), all of these rods are dropped into the core. The overall design provides for either
bank operation or individual operation of the rods. Two reactivity compensation rods (shim rods) working in pairs
provide for emergency protection of the second kind (junior scram). The emergency signals of the second kind in-
clude signals for exceeding the preset power level, for power supply outage, and for various other process signals.
37 boron carbide rods are designed for overriding poison burnout effects. These rods compensate 8% of the reac-
tivity and are used as on-off controls.
Power monitoring, control, and protection of the reactor is carried out by monitoring neutron flux with com-
pensated ionization chambers. These ionization chambers are placed in special pressure tight suspension supports
positioned in the space between the reactor well aid the reactor vessel. A total of 12 channels are available for
the suspended supports. Five lead-shielded channels are designed to accommodate startup suspension supports.
The control and protection system is based chiefly on magnetic amplifiers which assure stable performance
at low temperatures and easily withstand transportation over long distances.
Reactor control is performed by means of a single automatic control; a second stands on hot standby. A pre-
set power level can be held stable to within ?1%. An automatic startup control is available for starting the reactor
up automatically, bringing the reactor up from the nominal power level of 1 0-4 to 1O/0 up to a level of 1 to 1O'0
with a predetermined period. Over that range, the automatic startup control also provides protection by controlling
the rate of power rise.
Investigation of Dynamic Response and Control of Power Station
The dynamic response of power stations has been studied on an analog computer, and later directly on the
ARBUS power station. As a result of these dynamic investigations, it has been found that one special feature of this
power station is the comparatively slow unfoldment of processes under both operating and scram conditions. The
slowness of all the transients, accounted for by the comparatively large coolant volume in the primary loop and the
large volume of boiling water in the steam generators, is an advantage from the standpoint of the temperature re-
sponse of the fuel elements and the primary-loop components.
The study of temperature self-regulation of the power station showed that the power station exhibits adequate
stability for normal operation at a predetermined set of operating conditions without requiring an automatic pile
neutron flux control. The maximum permissible reactivity jump developed by the system in response to a permis-
sible deviation of the process parameters is roughly + 0.113 and-0.313.
1203
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
UI)
/
.
' ?
1
51........-
?
860
-zf'
./ a
/
6?"*".-
.
855
o
0
850
50
100
5.5
5,0
a)
4.5 k
150 E, MWh
Fig. 5. Temperature variation of gas oil viscosity:
1) original gas oil in loop; 2) gas oil containing
9.03% high-boiler radiolyzates.
The self-regulation operation, an automatic transition
of the power station for one power level to another with the
automatic controls shut off can occur given an appropriate
change in the temperature level of the coolant and in the
steam pressure in the secondary loop.
Investigations of perturbations in coolant flowrate have
shown that flow rate oscillations with amplitudes to 1011? and
frequencies of 0.01 Hz and higher are tolerable. No changes
in process parameters are observed at any frequency above
0.3 Hz.
When the station is operating at rated levels, an instan-
taneous drop of 260 kW in electric power load results in an in-
crease of 3 atm, , approximately, in the steam pressure in the
secondary loop, whereas a subsequent step-up of 260 kW in the
electric load will return the system parameters to their origi-
nal values.
It was found as a result of an investigation of a scram cool-down of the primary loop when one of the circu-
lating pumps malfunctioned that the temperature of the fuel element surface will not exceed the permissible maxi-
mum if emergency turbine pumps go into action nolater than 3 sec after the emergency signal becomes operative.
As direct experimentation conducted at the power station showed, this takes a time of 0.2 sec.
The off-peak operating time of the turbogenerator when working to satisfy the power
station itself will be 18 min when the reactor emergency protection system is actuated.
The results of analog computer studies were verifed by the direct power station tests.
ation in the power station parameters when the electric load was stepped up.
Maintenance and Cleanup of the Primary Loop
requirements of the power
Figure 4 shows the van-
The basic requirements imposed on power stations whose primary loop materials are carbon steels and which
require no biological shielding is the purity of the organic coolant employed. Care must be taken to keep the equip-
ment and piping of the loop free from contamination and from corrosion products in order to meet this requirement,
employing either mechanical or chemical means with subsequent protection with a volatile inhibitor (5011? aqueous
solution of monoethanolamine) and leakproofing during hauling and assembly operations. After assembly has been
completed and the loop has been pressurized with dehumidified air, the loop may then be filled with DA fuel (close
to gas oil in composition) with an iron content of roughly 1 mg/liter; this fuel is then used for hot blowdown of the
loop. The temperature is kept close to operating temperature during this operation. The operation is carried out in
three stages to maximize its effects. The fuel is drained out and replaced with fresh fuel for each such stage. The
blowdown process is monitored by checking the iron content in the fuel.
The blowdown and purification technology so described made it possible to achieve a startup of the power
station with an iron content of 0.2 to 0.3 mg/liter in the gas oil.
Coolant Regeneration
A specially designed regeneration system for treatment of the organic coolant by continuously bleeding a part
of it to a hydrogenation reactor was resorted to in order to remove polymers and unsaturated compounds from the
ARBUS primary loop. In this reactor (hydrogenizer), hydrogenation of unsaturated unstable compounds and degrada-
tion of polymers with the formation of compounds similar to the original compounds in their physicochemical prop-
erties was achieved under hydrogen pressure on an aluminum-cobalt-molybdenum catalyst bed, with up to 80% re-
covery yields. The remaining 20% consisted of lightweight off-products and coke. An additional purification of
the coolant from traces of metals and sulfurs present in it also occurred in a similar regenerative process. The reader
should note that the hydrocracking step is facilitated in this instance by the nature of linkages in radiation-chemi-
cal polymers. The regeneration conditions were so chosen that aromatic compounds would not undergo hydrogenation.
As studies have shown, the optimum process parameters for regeneration by hydrogenation of coolants fabri-
cated from gas oil petroleum cuts are: hydrogen pressure 40-60 atm, reactor temperature 350?-380?C, coolant
1204
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified
5
Viscosity v, centistokes
3
2
1
and Approved For Release 2013/03/14
14
v
0 40 80 120 160 200 t ?C
Fig. 6
: CIA-RDP10-02196R000600120002-4
/4
12
-
2 ./*
?
1
.50
100
10
8
6
4
2
Iodine number
150 E, MWh
102t
8.1021 9 1021
Radiation energy absorption, eV/g
Fig. 7
Fig. 6. Variation in viscosity and in specific weight of gas oil as a function of integrated exposure
dose: 1) viscosity; 2) specific weight.
Fig. 7. Accumulation of high-boilers and variation of iodine number, as functions of the integrated
exposure dose: 1) concentration of high-boiling radiolyzates; 2) iodine number.
Composition of Gas Formed in Radiolysis
Components
Contents
W/O
v/o
Hydrogen
26.670
83.077
Methane
24.191
9.415
Ethane-ethylene
13.634
2.829
Propane
11.079
1.566
Propylene
9.060
1.343
N-butane
5.496
0.590
Acetylene
0.548
0.132
Butane
2.081
0.224
Allene
0.386
0.061
B, i+ a-butylenes
6.219
0.692
3-Butylene + divinyl
0.636
0.071
mixture becomes separated
volume rate 0.5 ICI, and molar ratio of raw material to hydro-
gen from 1:5 to 1:10. The basic layout of the coolant re-
gneration system is seen in Fig. 1. Primary-loop coolant flows
(at 200 to 250 liters/h) through a metering pump. The meter-
ing pump sends the gas oil under 45 to 60 atm pressure down-
stream to mix with a counterflow of hydrogen circulating in
the system and obtained by electrolysis of water in the elec-
trolyzer, whence, the hydrogen is fed to the system by a boost-
er diaphragm compressor. The gas oil and hydrogen mixture
is heated in a regenerative heat exchanger, and then becomes
heated to operating temperature in an electric furnace. The
mixture then arrives subsequently in the catalyst-packed re-
actor. The mixture of hydrogen and regenerated gas oil
emerging from the catalyst-bed reactor gives up its heat in
the heat exhanger and ends up cooled down to a temperature
of 30-50?C in the cooler. Downstream of the cooler, the
in a gas separator unit, whence, the gas oil, on passing through the cermet and felt fil-
ters, gains access to the primary-loop make-up tanks, while the hydrogen is recycled to the circulating compressor.
As a result of the fact that gaseous polymer degradation products (methane) are formed in the hydrogenization
regeneration process, slight quantities of the circulating gas are continuously vented to the stack. The total flow-
rate of hydrogen is 5 normal m3/h, of which about 4 normal m3/h are delivered directly to the reaction.
Radiation-Chemical Coolant Modifications
The radiation-chemical characteristics of gas oil stabilized by hydrogenation obtained at the ARBUS nuclear
power plant appear on the whole to be in excellent agreement with the results of preliminary experiments and loop"-
test experiments. Figure 5 shows the temperature variation of the viscosity of the original primary-loop gas oil and
the irradiated primary-loop gas oil in the ARBUS plant, with a content of 9.03% high-boiling radiolyzates. Figures 6
and 7 show the variation in the density, viscosity, iodine number, and content of high-boiling radiolyzates as the
integrated exposure dose is increased and with the regeneration set out of operation. Figure 7 also plots the energy
radiation absorption scale per gram of gas oil.
The initial molecular yield of high-boiling products was computed on the basis of the data obtained, and
found to be ?2 mole/100 eV. This was later reduced to ?0.5 mole/100 eV with long-term operating experience.
1205
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
The table lists the compositions of gases formed during the radiolysis. The coolant flowrate was 20 to 30 tons
per year to replace losses by radiolytic decomposition at 10010 power load.
SUMMARY
The installation and the experimental operation of the ARBUS nuclear electric power generating station dem-
onstrated the feasibility of constructing organic-cooled nuclear power stations in remote regions of the USSR. The
correctness of the calculations and principal design decisions; the realizability of equipment fabrication and of the
carbon steel primary-loop piping without any biological shielding, and the possibility of exploiting available mass-
produced petrochemical equipment and standard fittings, giving due attention to the specific requirements of a nu-
clear facility, were confirmed in operation. This nuclear power station exhibits adequate operational stability, with
the added advantages of simplicity and reliability in operation under a variety of operating conditions.
In subsequent work, improvements can be made in the engineering costs of power stations of this type, by in-
creasing the process parameters (employing regenerable coolant of higher thermal stability), and by introducing other
power plant improvements based on the accumulated operating experience.
1206
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TES-3 COMPACT ATOMIC POWER STATION ?
N. M. Sinev, A. K. Krasin, I. F. Bychkov,
0. I. Blokhin, D. L. Broder, V. N. Gabrusev,
Yu. V. Dudnikov, V. A. Zhil'tsov, M. A. Koptev,
A. Ya. Komarovt, A. P. Kotov, M. N. Lantsov,
G. A. Lisochkin, G. A. Merzlikin, I. G. Morozov,
Yu. I. Orekhov, Yu. A. Sergeev, P. N. Slyusarev,
G. N. ,Ushakov, N. V. Fedorov, V. Ya. Chernyi,-
and V. M. Shrnelev
Translated frorn Atomnaya Energiya, Vol. 17: No. 12,
pp. 448-452;December, 1964
Introduction
In correspondence with the plan for the development of nuclear power in the USSR, the TES-3 mobile large-
unit atomic power station with a water-moderated water-cooled reactor was set in operation on an experimental
basis in 1961. TES-3 constitutes a full-scale demonstration experimental-low-power atomic power station (APS)
designed for operation in those regions of the USSR where the use of small APS .is economically expedient. For such
APS, it is especially important that the construction and assembly work at the site of operation be kept at a mini-
mum. In this regard, TES-3 constitutes an experiment in producing APS which would leave the factory n a state
of maximum readiness for operation. Virtually all the equipment of TES-3 is assembled in four large units and
mounted on four self-propelled caterpillar-track flatcars with heated coach-type bodies. This-makes it possible to
operate TES-3 without constructing special buildings, while the preparation of the site mainly consists in providing
the biological shield.
It is planned to utilize TES-3 mainly as a stationary plant without moving it to another site. Therefore, a
number 'of units connected with the drive of the self-propelled power cars can be dismantled after the station reaches
the site of operation (diesels, transmission gear, tracks, etc.). In this case, the chassis of the cars play the role of
ordinary foundation frames, while the space which becomes available can be used for auxiliary rooms.
Fig. 1. General view of the TES-3 site.
'Report No. 310, presented by the USSR at the Third International Conference on the Peaceful Uses of Atomic
Energy, Geneva, 1964.
f Deceased.
1207
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 1. Values of the Critical Height h and of dp /dh for Cores '
with Different Numbers of Fuel Assemblies
Number
of fuel
assem-
blies*
Core di-
ameter
D, cm
D/h
h, cm
dp/dh, X 102 cm
experi-
ment
calcu-
lation
experi-
ment
calcu-
lation
31
40.4
0.69
58.6
49.2
0.173
32
41.1
0.75
54.6
47.0
0.204
36
43.4
0.99
44.0
40.8
0.350
0.303
60
55.7
1.96
28.4
28.7
0.999
1.006
74
61.9
2.46
25.1
26.3
79
63.9
2.64
24.2
25.8
1.36
1.444
*The central cell does not contain a fuel assembly.
Fig. 2. View of the self-propelled power cars with the
control panel and the turbogenerator. The car tracks are
covered for heating purposes.
k/k
a
0,0
40
54111
A'
ki
1
Eill
El
50
100
150
200
250
Fig. 3. Dependence of the reactor's reactivity .ak/k on
the mean water temperature t in the reactor (the solid
curves represent the experimental results, while the dashed
curves denote the calculation data). 1) Reactor with 74
fuel elements; 2) same (the core contains two manual-
control rods); 3) same (the core contains six manual-con-
trol rods); 4) hot critical assembly with 85 fuel assemblies.
1208
If necessary, the communications between the
self-propelled units can be dismantled and the station
transferred to another site. The weight and the over-
all dimensions of the self-propelled power cars make
it possible to transport them by railway. During trans-
port, the reactor is cooled by means of a radiator pro-
vided in the reactor car. The replacement of depleted
fuel assemblies can be performed under field condi-
tions without removing the reactor lid by using a re-
loading container and a 25-ton self-propelled crane.
Description of the Station
The TS-3 power station was installed at Obninsk
at the site of the first atomic power station in the world
in such a manner that the preparation of the area, the
assembly, and the operation of the station corresponded
most closely to these operations as they would be per-
formed under field conditions. The TBS-3 site (Fig. 1)
comprises the self-propelled power cars, a storehouse
for spare parts, and a small prefabricated building
housing the turbogenerator panel, a transformer for the
station's own needs, and the radiation health checkpoint.
The self-propelled power cars with the reactor
and the other first-loop equipment are located in a
common trench with a depth of 2.8 m, whose walls
and cover are made of prefabricated reinforced-con-
crete sections and are covered on the outside with soil,
which constitutes the basic biological shield. The self-
propelled power cars with the turbogenerator and the
control panel are situated at the ground level (Fig. 2).
TES-3 constitutes a two-loop steam turbine
plant. It is provided with sectionalizing and emer-
gency arrangements for reliable self-contained oper-
ation of the station. The station is automated; its
power level varies automatically with changes in the
turbogenerator load. The basic characteristics of the
power station are:
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Generator power, kW
1500
Reactor power, kW
8800
Pressure in the first loop, atm abs
130
Water temperature, ?C:
at the upstream end of the reactor
270
at the downstream end of the reactor
300
Parameters of the second loop:
steam generator pressure, atm abs
20
steam superheating temperature, ?C
280
pressure in the condenser, atm abs
0.13
Cooling water discharge, t/h
1000
Weight of the station's equipment, t
210
Weight of transported biological shielding,
included in the above, t
28.5
Weight of all self-propelled cars, t
310
Duration of run, days
250
Reactor Physics
The physical characteristics of the TS-3 reactor were determined on the basis of the two-group theory by
taking into account the multiplication of neutrons in moderation; they were checked by means of three critical as-
semblies which were put together successively. At the initial stage, the experiments were performed on a critical
assembly that only approximately simulated the core of the TS-3 reactor. The subsequent experiments were per-
formed by using two full-scale critical assemblies, which simulated the TS-3 reactor to the full extent. "Cold"
experiments for checking the design of the control elements and devices for charging the reactor with uranium were
performed by using one of these assemblies. It was possible to heat the last, "hot" critical assembly almost to the
calculated temperature, which made it possible to investigate the temperature effect.
The initial reactivity was found by using the dependence p =f(dp/dh). The critical height h was
varied by changing the number of fuel assemblies (Table 1).
The thus obtained critical mass of the reactor with the calculated core, i.e., with 11 fuel assemblies removed,
differed from the theoretical value by 5%. The reactivity of the cold uncontaminated reactor (0.114) also was in
satisfactory agreement with the calculated value, which was equal to 0.123.
T he tern per at ure effect of tea ct iv ity determines to a large extent the safety of the reactor and
convenience in operation. Therefore, a thorough investigation of this effect was carried out in experiments on all
critical assemblies and during the actual start-up of the reactor. The experiments showed that the temperature ef-
fect is highly sensitive to the presence of absorbing rods in the core and to the withdrawal of individual fuel assem-
blies from the core (Fig. 3).
The compensating ability of the control elements was investigated for rods made of different
materials, which were characterized by different designs and dimensions and were arranged within fuel assemblies
and in individual cells. The data pertaining to the chosen control system are given in Table 2.
Biological Shield
The biological shield of TS-3 consists of two parts: the immobile part, which is provided at the operation
site, and the transportable part. The transportable part of the shield is made of lead and is installed in the tank of
the biological shield. The tank, which is provided with a coiled pipe for cooling, is filled with distilled water or
a solution of boric acid. The thickness of the lead layer is 100-190 mm. The water (-700 mm) serves as the anti-
activation shield for the tank walls, the coiled pipe, and the structure of the self-propelled car. Before transportation,
the tank is emptied.
A specific feature of the biological shield of TES-3 is the use of lead as the heavy component, which, in its
pure form, does not acquire the long-lived induced y- activity; this component was inserted directly in the tank of
the biological shield without jacketing. Another characteristic of the shield is the use of a boric acid solution, which
also reduces the activation of the materials in the shield and the structure of the self-propelled reactor car. The
1209
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 2. Compensating Ability (Ak/k ? 103) of the Control Elements
Control element
Material
t = 20?C
t= 280?C
experiment
calculation
experiment
calculation
Manual-control rod
B4C
6?0.16
5.35
12?0.3
8.97
Automatic-control rod . . .
Boron steel (2% boron
by weight)
1.51?0.04
1.78
3.4? 0 (:18
3.05
Emergency rod
B4C
5.7?0.15
4.84
6.9?0.18
5.72
"Soft" compensating system
Stainless steel
114? 2
90
?
?
G,t/h, Nel, kW
15 150
10
'Cl
5 5 G,t/h G,t/h
8 8
CL 6 6
II 20 40 60 60 11/0
4 4
2 4 6 min 0 2 4 5 min
Fig. 4 Fig. 5
Fig. 4. Dependence of the steam discharge G (1) and of the generator power Nei (2) on the reactor power
level. ) Measurement; --) calculation.
Fig. 5. Reactor operation under self-control conditions (a and b?cases where the water supply discharge was
increased and reduced by 5 t/h, respectively), 1) Water temperature at the upstream end of the reactor; 2)wa-
ter temperature at the downstream end of the reactor; 3) water supply discharge.
v
9
?
2.,
/
/
/
/ 1
G
/
/
/
/
/
/
/
-
t,?C
290
286
282
278
274
290
286
282-
278
274
3
Reactor power level, %
TABLE 3. Radiation Level at the Lateral Surface
of the Self-Propelled Reactor Car, AR/sec
Exposure
time, days
Experiment
Calculation
0.1
30.2
33.2
0.5
6.92
10.4
1.0
3.02
4.36
10
0.195
0.338
20
0.087
0.279
30
0.055
0.187
Operation of the Power Station
effectiveness of the above measures for reducing the activation
y-radiation was confirmed during the operation of TES-3. The
biological shield was calculated in the rnultigroup P1- approxi-
mation, where the neutron group with E 1.5 MeV was assigned;
the calculations were checked by measuring the neutron distri-
butions in different shield variants for the critical assembly.
Table 3 provides the y-radiation levels at the lateral sur-
face of the self-propelled reactor car, measured at the initial
period of TES-3 operation. Before the measurements were per-
formed, the reactor was operating at a power level of 3000 kW
for 20 days. The shield tank was filled with distilled water.
The purpose of the operation of TES-3 was a check of its operating ability under different conditions and an
investigation of the problems arising in designing and operating plants of this type, which, in the final analysis, will
make it possible to determine the advisability of utilizing low-power APS in the national economy of the country.
Figure 4 provides data characterizing the operation of the station at various power levels. Figure 5 shows the results
of one of the experiments on reactor self-control. Under conditions of self-contained operation, the power station
is started up by means of a 150-kW starter generator, which is driven by the propulsion diesel of one of the self- pro-
pelled cars. The power station operates under different load conditions: at constant power and under conditions of
1210
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 4. Yield of Radiolytic Gases, norm. liter/h ? m3 sudden changes in the turbogenerator load. Emergency
shut-downs were simulated several times, in which case
the TS-3 units and systems were first supplied from a
storage battery and then from a diesel generator.
of Solution
Concentration
of boric acid,
Reactor power level, MW
3.3
5.5
/0 by weight
H2
02
H2
02
0.04
0.35
0.2
0.68
0.42
1.7
1.45
0.8
3.0
1.45
3.5
3.7
1.8
5.6
2.55
Water Cond it ions. Ammonia or hydrazine hydrate
were added to the water in the first loop for binding dis-
solved oxygen and increasing the pH to its optimum value
(9-10). Hydrazine hydrate proved to be much more ef-
fective. After it was introduced, the oxygen content in
the first loop was equal to 0.005 mg/liter. Hydrazine
hydrate was also used in the second loop, where it re-
duced the oxygen concentration in the feed water to 0.01-0.005 mg/liter. In order to control the corrosion of lead
and steel in the biological shield tank, the pH value in the boric acid solution was maintained at a level of ?7 by
means of ammonia. Table 4 provides data on the radiolysis of water in the biological shield tank for different con-
centrations of boric acid and different reactor power levels.
Conclusions
The construction and operation of the TES-3 plant have shown that the experiment of producing a large-unit
mobile power station with a water-moderated water-cooled reactor was entirely successful. The prolonged operation
of TES-3 confirmed the reliability, satisfactory controllability and convenient servicing of this type of power station.
At the same time, the operation of TES-3 has shown that further improvements are possible, in particular, a
greater degree of automation, an increase in the run duration to two or three years, a switch to natural coolant cir-
culation in reactor cooling, etc.
We should also mention the satisfactory agreement between the theoretical basic parameters of the power sta-
tion and those actually obtained, which was largely due to the many experiments performed at the design stage.
1211
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
PHYSICAL AND OPERATING CHARACTERISTICS OF THE SM-2 REACTOR*
S. M. Feinberg, N. A. Dollezhal', E. D. Vorob'ev,
V. A. Tsykanov, I. Ya. Emel'yanov, V. M. Gryazev,
A. S. Kochenov, Yu. M. Bulkin, V. I. Ageenkov,
and P. G. Aver'yanov
Translated from Atomnaya Energiya, Vol. 17, No. 6,
pp. 452-463, December, 1964
INTRODUCTION
The SM-2 reactor was designed for extensive research in nuclear physics, solid state physics, metallurgy, ra-
diochemistry, reactor construction physics and engineering, and in many other branches of science and engineering.
The core is a parallelepiped, 42 x 42 X 25 cm in size, with a cavity measuring 14 x 14 x 25 cm (neutron "trap").
The corners of the parallelepiped, measuring 7 X 7 x 25 cm, are occupied by compensating rods. The core is made
up of fuel element assemblies, 7 x 7 X25 cm in size, and is surrounded by a reflector made of beryllium oxide blocks
with water interlayers. The fuel element assemblies contain 54 fuel element plates 0.8 mm thick which are sepa-
rated by water-filled gaps 1.65 mm wide.
There are five horizontal, one sloping, and 18 vertical channels in the reactor for performing experiments.
The physical features of the reactor and its basic parameters were discussed in [1]; the fundamental engineering
decisions were discussed in [2]. The SM-2 reactor is the first intermediate research reactor with water moderator.
Construction of the reactor was completed in 1961. Physical startup of the reactor was achieved in October,
1961. From November, 1962, the reactor has been operating at nominal parameters. The maximum power reached
by the reactor was 55 MW.
After a period of reactor operation at reduced power and particularly at planned parameters, an extensive
program of scientific research work was carried out which dealt both with the study of the reactor itself and its sepa-
rate units and also with the completion of a program of irradiation of various materials in the experimental channels.
In addition, operation of the reactor during the time mentioned allowed a study of the efficiency of its basic
components.
The idea of building the SM-2 intermediate research reactor and the basic construction design came from
S. M. Feinberg. He also directed the entire complex of operations involved in construction of the reactor. The
design of reactor construction was carried out under the direction of N. A. Dollezhal' and Yu. M. Bulkin. Experi-
mental studies in the initial stages of the project were made by E. D. Vorob'ev, V. B. Klimentov, and V. M.Gryazev.
Several physical computations, consisting of a mathematical program of P1- approximation reactor calculations, were
completed during this stage by I. K. Levin and N. Ya. Lyashchenko. During the second stage of the project, experi-
mental studies of reactor physics were continued with the participation of V. A. Tsykanov and others. The immedi-
ate assistant to the scientific director of the project was V. A. Tsykanov who did much toward the development of
the engineering plans for the reactor and who directed the construction and operation of the reactor in the later
stages of the project. Construction of reactor control and protection systems was worked out under the direction of
I. Ya. Emel'yanov. Starting with the second stage, the basic physical calculations were made by A. S. Kochenov.
The fuel elements for the reactor core were developed in accordance with the suggestions and techniques of V.I.
Ageenkov. N. G. Aver'yanov and other members of the operating staff played an important part in mastering the
intricacies of operating the SM-2 reactor.
*Paper No. 320, presented by the USSR at the Third International Conference on the Peaceful Use of Atomic Energy,
Geneva, 1964.
1212
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 1. Critical Masses for Various Core Configurations
Type of system
Channel conditions
Core con-
figuration
as shown
in Fig. 1
Beryllium oxide assemblies installed
in central cavity
Water in central cavity
The same
Beryllium inserts and water
in central cavity
*Reduction of the number of fuel eleme
T Reactivity was computed for a system
Channels open
Vertical channels closed
with Be? plugs
Lead plug in horizontal
channel III. Be0 plug
in horizontal channel V
The same
Channels open
a
the system s
of 20 fuel element assemblies.
nt assemblies by one makes
Number
of fuel
element
assem-
blies*
Reactivity, 010
U23 5 ,
kg
meas.
calc.
13
8.6
0
20
13.2
0.5
1.8
20
13.2
0
1.1
21
13.9
0.7
0.8T
22
14.5
0.7
18
11.9
1.0
ubcritical.
A large group of scientists, builders, engineers, and workers participated in the work of planning, construction,
and operation of the SM-2 reactor; the authors wish to express their gratitude to all of them.
PHYSICAL CHARACTERISTICS OF THE REACTOR
The linear dimensions of the SM-2 reactor are commensurate with neutron migration length, but the core and
reflector have a complex geometry. Therefore, reliance on physical calculations for reactor planning was risky,
and the main consideration was given to experiments.
The first experiments have already been described [1]. Basically, the experiments were performed with a
homogeneous core which contained construction materials in addition to uranium and moderator. Critical experi-
ments performed with circular cores and water reflector were inadequate. Because of the lack of uranium in these
experiments, one could not manage to introduce the required amount of construction materials into the core. Sev-
eral critical experiments were performed with a uniform core and a beryllium oxide reflector which covered only
a small portion of the lateral surface of the core.
On the basis of this clearly unsatisfactory information, it was expected that the critical loading for the SM-2
reactor would be 6.8-7.3 kg. Furthermore, the loss of reactivity because of the experimental channels was not taken
into account. Subsequent experiments, in which the core and reflector of the SM-2 reactor were completely repro-
duced, showed that the critical loading was considerably greater.
Appropriate computing techniques were developed simultaneously with the experiments. It was determined
that Pr approximation calculations [1] (with 12 energy groups) for the distribution of neutron flux throughout the
core and "trap" agreed satisfactorily with experimental data. However, the magnitude of the critical loading was
in satisfactory agreement with experimental values only in the case of one-dimensional core and reflector geome-
try. It was clear that a physical computation of some core configurations could only be made with a two-dimen-
sional approximation.
A calculation of the neutron spectrum in the core indicated that the main absorption of neutrons (-9001o) was
below an energy of 10 eV. The age of the thermal neutrons is 35 cm2, and the energy region from 10 to 0.1 eV
contributes a total of 2 cm2. Therefore, neutron leakage from the reactor is determined by the energy region
above 10 eV.
1213
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
3
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
?
-0
2
3
0
0
0,0
0
0
?
0
?
0
0
0
?
a
0
0
?
0
2
3
4
3
2
-oo
-0
0
?
0
0
0
0
0
00
0
0
0
?
0
0
?
?
3
2
2
3
4
5
c (MCC
14141414104
411111 D
MP=
1A1A1A.0.40AU
['You E
0
0
0
0
0
Fig. 1. Core configurations: 1) compensating rods;
2) removable Be0 assemblies; 3) fuel element assem-
blies; 4) water; 5) beryllium inserts; 6) reflector;
7) safety rods.
1
In this connection, it is suitable to assume the following two-group model for neutron physics calculations:
the fast neutrons are not absorbed but undergo considerable migration; only slow neutrons, whose migration is small,
(with energies less than 10 eV) are absorbed.
To solve the reactor equations in the two-group, two-dimensional approximation, the mathematical program
put forth in [3] was used. The SM-2 reactor calculations were carried out for both plane and cylindrical geometries;
In the plane geometry case, neutron leakage in the vertical direction was taken into account by assuming that the
group absorption cross sections Ei were
,
= ?
(1)
Here, Er is the absorption cross section in an infinite medium for the i-th group; Di is the diffusion coefficient for
the i-th group;(H+28 ) 2' where H is the height of the core; 6 is the effective correction.
Neutron leakage through vertical gaseous channels was taken into account in the following manner: all mac-
roscopic cross sections were assumed to be zero in the channel volume, and the diffusion coefficient
Di = t3r_i (2)
where dh is the hydraulic diameter of the gaseous channel; Xir i is the transport length for the i-th group in the
region surrounding the channel.
The effect of the horizontal channels on reactivity was taken into account by means of experimental corrections.
For the cylindrical geometry, only the neutron flux distribution along the height of the reactor was computed.
The effect of the experimental channels was not considered.
Using the two-group, two-dimensional approximation, calculations were made of the reactivity with different
core configurations, of the control rod compensating capabilities, the reactivity changes with U235 burnup, and of
the distribution of fast and slow neutron fluxes. The calculations were in satisfactory agreement with experiment.
1. Critical loading
Critical loading was investigated for various core configurations: with a central water-filled cavity, with
beryllium inserts placed in the water-filled cavity, with a central cavity filled with beryllium oxide, etc.
The experiments were performed on a special test stand Ind directly in the reactor vessel. The critical state
was achieved by gradually filling a previously assembled core with water. The results of several experiments to
determine the critical mass of the reactor are given in Table 1.
1214
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 2, Compensating Rod Reactivities
Reactivity, 19
Compen-
sating rod
measured
calculated
20 assem-
22 assem-
23 assem-
20 assem-
blies, Fig, le
blies, Fig. le
blies, Fig. lg
blies, Fig. lc
I
1.11
0.81
0.81
1,15
II
143
1.53
1.82
1.15
III
1.11
1.35
1.25
1.15
IV
1.05
0.77
0.71
1.15
TABLE 3. Effect on Reactivity of Various Materials
Located in the Peripheral Vertical Channels
(see Fig. 1g)
Channel condition
Reactivity change, 10
channels
2 and 3
channels
4 and 5
Gas replaced by water
?0.03
?0.08
Channel lining (steel pipe
69 mm in diameter,
3 mm wall)
+0.04
+0.18
Be0 plug 62 mm in
diameter
+0,1
+0,58
Closely packed bundle
of nickel rods
?
+0.04
Loaded with 170 g of tg35
+0,23
?
TABLE 4. Effect of Reactivity of Various Materials
Located in the Horizontal Channels
Reactivity change, lo
closely
channel
Channel condition
packed
directly
bundle
adjoin-
of nick-
irig the
el rods
core
_
Be() plug 62 mm in diameter
and 300 mm long
+0,1
+0.43
Water plug 73 mm in diam-
eter and 260 mm long
+0,04
+0,22
Plug of lead discs 72 Trim in
diameter and 8 mm thick
with plastic spacers giving a
total length of m mm
+0.1
+0.43
The critical mass for the actual construction with a water-filled cavity at the center and without experimen-
tal channels turned out to be 11 kg in contrast to the preliminary estimate of 6,8,7,3 kg mentioned above.
The experimental channels increased the critical mass to 13,5 kg. In order to reduce it, beryllium inserts
were installed in the neutron trap. This lowered the critical mass to 8.6 kg without experimental channels, and to
11,2 kg with them.
A physical calculation of the two-group, two,dimensional type underestimated the value for the critical
loading. However, the difference from the experimental values did not exceed 100/o. below are given values for
the reactivity produced by replacing the beryllium oxide assemblies by fuel element assemblies for the cote con-
figuration shown in Fig, lg:
Cell
A-2
4-5
B_ 1
B-5
B-6
C-4-.2
C-5
C-4
F-4
Reactivity, 0/0W .
0,67
1,05
0,56
1,13
1,08
1,05
1,57
1,29
1,22
As should be expected, it is clear from the results that the effect of fuel element assemblies on reactivity
depends on core configuration, position of assembly insertion, and the presence of experimental channels nearby.
2. Reactivity compensation
The compensating capabilities of the control rods were determined by means of reactor excursions. The re-
activity was calculated by the "inhourn formula, On the basis of special calculations, the ratio i3eff/6 was assumed
to be 1.4 for a cote containing 20 fuel element assemblies, and 1.3 for 28 fuel element assemblies.
415
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
The reactivity of the compensating rods which were
located in the corners of the core is given in Table 2 for
three core configurations, from which it is evident that the
reactivity of the four compensating rods was ,-.4.5% (inter-
ference between these rods was not observed). For normal
operation of the reactor, this was insufficient. Therefore,
to strengthen the control and protection system, four safety
rods were located in the beryllium inserts installed in the
neutron trap in addition to the four compensating rods.
These rods consisted of a beryllium ejector and a water-
filled cadmium tube (1 cm in diameter). While the re-
actor was running, the absorbing part of the safety rods
10 15 20 25 was located above the core and consequently did not per-
turb the neutron flux. The compensating capability of a
single rod was 1%. Because of interference, the reactivity
of the four rods was 2.8%.
The four compensating rods which were positioned
according to plans in the body of the reflector had little
reactivity. Therefore they were removed from the reactor,
and the channels thus released were used for experimental
purposes.
1;CM
Fig. 2. Distribution of U235 fission density along a re-
actor radius (solid curves were calculated by the two-
group, two-dimensional approximation; experimental
values in the "trap" were increased 1.3 times): 3) wa-
ter in trap; 2) 25 mm thick Be0 plates in trap along-
side core; 3) beryllium inserts in trap.
The total reactivity of the compensating and safety rods was 6.6% with a core loading of 28 fuel element
assemblies.
3. Effect of Experimental Channels on Reactivity
The effect of the experimental channels on reactivity was studied on a test stand. It was determined that
loading the central channel (water-filled) with material having small absorption cross sections (aluminum, lead,
beryllium oxide, beryllium) increased the reactivity. Beryllium and beryllium oxide had a strong effect in the in-
crease in reactivity. For example, inserts of beryllium installed in the neutron trap between the central channel
and the core increased reactivity by 5%. Six cylindrical slugs with 253 enrichment and a diameter of ?1 cm with
a total U235 load of 18 g uniformly distributed over the cross section of the central channel increased the reactivity
by ?1%. Ten SM-2 fuel element plates installed with a 2.45 mm spacing increased the reactivity by ?070.1
The effect on reactivity produced by various materials located in the peripheral horizontal and vertical chan-
nels is shown in Tables 3 and 4. Both tables refer to the core configuration shown in Fig. 1g. The outer peripheral
channels have practically no effect on reactivity. The total loss of reactivity resulting from the presence of experi-
mental channels in the external reflector amounts to ?4%.
4. Space-Energy Neutron Distribution
Experimental studies of neutron spectra and of the spatial distribution of neutrons over the reactor volume
have been carried out with miniature fission chambers containing U235 and with indium, uranium, and gold detectors.
The diameter of the fission chambers was 1.4 mm, making it possible to make measurements in the gaps between
fuel element plates. The precise displacement and location of a fission chamber was accomplished with the aid of
a special measuring bridge. The U235 fission density in the experimental channels was measured with chambers of
larger diameters (to 5 mm). Results for measurements of the quantity Q = f c13(u) o5f (u)du are shown in Fig. 2.
Solution of the transport equation indicates that there is a considerable depression of the neutron flux in the
fuel element. The ratio of the value in a water gap Qw to the average value at the center of a fuel element Od
is ?1.3. Since the ratio Qt/O-d (where Qt represents the maximum flux in the trap) enables one to relate the maxi-
mum thermal flux to the specific power of the reactor, the ratio Qt/Qd, and not Qt/Qw, is of most interest.
Curves for the distribution of fission density, which was measured with U235 chambers, are given in Fig. 2.
The points give experimental results in the neutron trap multiplied by 1.3.
Experiments indicated that the coefficient of variation for heat release throughout the core was ?3.
lAs in Russian original text-Publisher.
1216
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
j
1,0
fiCa ; Ritt,
$
4 6
L4
0
It FA
IF
16
-14 -7
7 14
Fig. 3. Experimental distribution along a "radius" of the reactor
for activation by neutrons above the cadmium cutoff (Ncd) and for
the cadmium ratio (Rcd) in indium and gold; 1) NL1; 2) RAcd;
3) Nd; 4) 12.11d; 5) wall of central channel; 6) beryllium.
Eimolmumumiumi
111?1111111111111111111111111111111111111
mommirimppr.ffiinnummemilum
11111114M1MhZHNIIIIIIIIHIll
.-:,==1:4411===ranntr.e*.rxmin
EaVainiara=2;;;17:14:11Eng
annmsemsmiaboilm...111
1111111111n1111INIIhmillulii
MIE411111111E1K1111111111111i1111111
mommuminetimomisnminom
111111110111,1111111111111111111111111
mazarrir.-antan...r.rmarana
Num. IMMINIIIIrmuummunsaniou
11101111?Fli Emmmemilimommuni
1111117411111111111111111111111111111111111111
11111111111/11111111111111111111111111111111MMIIIHI
1114 111111111111111111111111.111111
1 10
Fig. 4
3000
2000
1000
-20
?10 0
Fig. 5
10 20 30 h, cm
Fig. 4. Energy dependence of neutron flux (13(z) and of flux along the energy axis, j(z); 1) 4'(z); 2) j(z);
z =E/kT, T =323"K.
Fig. 5. U235 fission density distribution along vertical channels: 1) channel 14; 2) channel 4; 3) chan-
nel 15 (nearby compensating rod withdrawn); 4) channel 15 (nearby compensating rod withdrawn approxi-
mately_halfway).
The radial distribution for density of activation by neutrons above the cadmium cutoff and for the cadmium
ratio with indium and gold is shown in Fig. 3. Indium and gold foils 0.1 and 0.2 g/cm2 thick, respectively, were
used in the experiments.
The calculation of neutron fluxes was performed in various approximations. The space-energy distribution
of the slowing down neutrons and the curves for fission density, which was measured with U235 chambers, were com-
puted in the P1-approximation. The fission density curves were also calculated by the two-group, two-dimensional
approximation.
1217
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
25
20
15
10
5
G, cm
Fig. 6. U235 fission density distribution in the neutron trap with insertion of
SM-2 fuel element plates; 0) no plates; A) three plates, 2.45 mm spacing;
A) six plates, 7.35 mm spacing; x) eight plates, 4.9 mm spacing; 0) 14 plates,
2.45 mm spacing.
The slow neutron spectrum was calculated for a homogeneous medium by means of the equation [4]
ze?z
z cW-d (t z'e?e)
dz'a (z") (z") dz" const]
+
V 2
1+ a ( ee?z.)
0 0
(3)
where z=E/kT; T is the medium temperature; Es(z) is the macroscopic scattering cross section; Ea(z) is the macro-
scopic absorption cross section; g is the mean logarithmic energy loss for neutron scattering by a free, stationary
nucleus; y is one half the ratio of the mean logarithmic squared energy loss to the mean logarithmic energy loss
for scattering by a free, stationary nucleus.
Equation (3) was solved on an electronic computer assuming that E5=g const and y =1. The relation for the
variation of absorption cross section as a function of energy was selected in accordance with the absorption cross
section for U235. The neutron spectrum was computed in the 0-2 eV energy range. At the same time, a computa-
tion was made of the slowing down flux along the energy axis, which is the fraction of neutrons absorbed below some
given energy, i.e.,
5 cy5,01) dE'
(E) _o
Scr5,41 dE'
(4)
The temperature of the medium was assumed to be 50?C. The energy dependence of the neutron flux cl.(z)
and of the slowing down flux along the energy axis is shown in Fig. 4. The calculated cadmium ratio for U235 from
capture was 2.44, and was 2.47 from fission. The experimental value for the cadmium ratio of U235 through fission
varied in the range 1.5-2.1 for the center of the core.
1218
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 5. Steady-State Xe135 Poisoning of the Reactor
I
1
Power,
MW
Number of
fuel element
assemblies
Average
power,
MW/liter;
Poisoning elk-k,7,,.
in the core
measured
calculated
4,1
20
167
2,03
2,00
28,0
28
818
3,77
3,55
33,7
28
985
3,69
3,70
Measurements of the fission density distribution
in the experimental channels were performed in the
system shown in Fig. lg. Vertical channels 2-7 and
horizontal channels I-V were filled with air, and ver-
tical channels 1.8-11 and 13-15 were filled with water.
The fission density distribution for U235 over height was
measured in vertical channels 4, 12, 14, and 15. The
results are given in Fig. 5.
Below is data on U235 fission density in the verti-
cal channels at the level of the center of the core:
Channel number
1
4, 5,
15
2, 3,
13
12
14
6
9, 10
8
7
11
-
Relative fission
density
1
1 0,11
0,067
0,057
0,045
0,034
0,030
0,027
0,017
0,013
I
During the measurements, all four compensating rods were inserted to the center of the core.
There is great interest in the neutron flux distribution in the central water-filled cavity when various amounts
of fissionable material are placed in it. A series of measurements was made to determine this; the results are shown
in Fig. 6. SM-2 fuel element plates were installed in a channel at various spacings. As can be seen from the curves,
placing 14 fuel element plates at a spacing of 2.45 mm in the channel leads to a reduction in thermal neutron flux
by a factor of approximately 20. The neutron spectrum in the vertical channels was measured by means of thres-
hold detectors (Au, In, Mg, Ti, Fe, Al). The neutron spectrum for several channels is shown in Fig. 7.
5. Steady-State Xe135 Poisoning of the Reactor
Data on the steady-state poisoning of the reactor by Xe135 are given in Table 5. Poisoning was calculated by
the formula
CO
?5 (VI + CrXe0 dE
P= 6/ 0
-5
Gc
axe(1) dE
(5)
where )'i+ y2= 0.064 is the total yield per fission of iodine and xenon fission fragments; X2 = 2.09 x 10-5 sec-1 is the
Xe135 decay constant.
Formula (5) embraces the following assumptions:
1) the nuclear density of Xe135 is considered constant over the core and is equal to a certain value which is
related to the average fission density;
2) the slow neutron importance is also constant over the core;
3) the slow neutron spectrum over the entire core is assumed the same as that within it.
In actuality, there is a spike in the neutron flux at the boundary of the core and this leads to the fact that the
nuclear density of xenon at that location can be greatly different from the mean value. In addition, the importance
and spectrum of slow neutrons at that position will be different from the corresponding quantities in the bulk of the
core. Consideration of the boundary effects mentioned is taken into account by means of perturbation theory. Vari-
ations in reactivity because of reactor poisoning are computed by the formula
k
A
k =OP'
where 0 is the fraction of neutrons absorbed in the uranium.
(6)
1219
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
(E 7(1 E
g9
0,8
g7
0,6
0,5
0,4
0,3
0,2
0,1
1
2
3
0 2 4 6
8 10 E, MeV
It is clear from Table 5 that the measured reactivity poisoning
at a power of 28 MW is greater than that at 33.7 MW. This contra-
diction points out the deficiencies in the experimental methods for
measuring reactivity, and indicates the error of measurement is not
less than 0.25%.
At a power of 50 MW, the Xe135 poisoning is close to the maxi-
mum and leads to Ak/k
6. Uranium Burnup and Fuel Assembly Replacement
The loss of reactivity during reactor operation is 2.2 ? 10-4%
per MW ? h. A calculation of the variation in reactivity by the two-
group, two-dimensional approximation assuming uniform uranium
burnup within the confines of an individual fuel element assembly
gives a value of (1.7-2.3). 104%/MW ? h. With this same approxi-
mation, replacement by fresh assemblies of four fuel element assem-
blies in the inside row of the core having 12.5% uniform burnup in-
creases the reactivity by 1.310, but the same replacement in the out-
side row increases the reactivity by 1%. With 25/0 burnup, these val-
ues are 2.8 and 2%, respectively.
Fig. 7. Integral neutron spectrum meas-
ured with threshold detectors (E is neutron 7. Effect of Temperature on Reactivity
energy, MeV); 1) channel 4 spectrum; Experiments to determine the effect of temperature on reac-
2) channel 19 spectrum; 3) channel 6 tivity have shown that reactivity monotonically falls with heating of
spectrum. the water in the primary loop while the temperature of the water in
the central channel is held constant. On the other hand, reactivity
increases when the water in the central channel is heated and the temperature of the water in the primary loop is
maintained constant. Simultaneous heating of the water in the core and in the central cavity leads to an increase
of reactivity up to ?30?C, and a fall in reactivity with further heating. The results of these experiments are shown
in Fig. 8.
8. Profiling Uranium Along Fuel Element, Assembly
One of the most complicated problems which come up in reactors of the SM-2 type is the correct profiling
of the uranium at the core boundaries.
Experiments performed with SM-2 fuel elements showed that they can operate with thermal loadings greater
than 107 kcal/m2 ? h, whith exceeds the maximum thermal loading in the reactor by a factor of two. Consequently,
the main purpose of uranium profiling along the core is not the reduction of thermal loading but the increase of
average burnup. Calculations indicate that, if the uranium loading in the fuel element assemblies is not profiled,
the average uranium burnup in fuel element assemblies removed from the core will be 51t, times less than attain-
able (k=1.3 is the variability coefficient for energy deposition with height in the core). This means, for example,
that if the attainable burnup is 35%, the average burnup will not exceed 6%.
The profiling assumed in the design (first plate containing 14 uranium, the second 24, the third and succeeding
ones 1) made it possible to increase burnup by 1.7 times. This occurred because the first plate with a complete
uranium loading was third from the inside boundary of the core where the neutron flux was less. Plates with lower
uranium content, i.e., with greater possible burnup, were located in high flux regions. During operation of the re-
actor, a new profile (0.25; 0.4; 0.6; 0.8; 1.0) was introduced which made it possible to raise the average burnup by
a factor of 2.5 in comparison with unprofiled fuel element assemblies.
All the data presented on burnup refers to a core without beryllium inserts. Placing beryllium inserts in the
neutron trap somewhat reduces the neutron flux at the internal boundary, and increase the average burnup by 1.3
times.
Thus, the measures taken made it possible to bring the average burnup toward the achievable value by a
factor greater than three. However, the gap between average and attainable burnup is still large and is ?2.
1220
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
0,2
0,1
0,1
0,2
0,3
2
--
_
i
. 31
35
% 41'.1
5 -N
55
60
65
70
75
t,?C
111.
\
4
3
1
-
Fig. 8. Effect of temperature on reactivity (Lip is
the change in reactivity, t is water temperature):
1) main loop water heated; 2) channel 1 water
heated; 3) simultaneous heating of main loop and
channel 1 water; 4) simultaneous cooling of main
loop and channel 1 water; 5) simultaneous heating
of main loop and channel 1 water with small vol-
ume of free gas in main loop.
3,0
2,0
1,0
3
I.
0 5 10 15 20 x,%
Fig. 9. Variation in thermal loading at fuel
element surface as a function of average U235
burnup in the fuel element assembly (without
beryllium inserts): q is the ratio of thermal
loading in a given fuel element to the aver-
age thermal loading in the fuel element as-
sembly; x is the average U235 burnup; 1, 2,
3, 4 are, respectively, the boundary, second,
third, and fourth plates.
Profiling the uranium in a fuel element assembly
increases the critical loading. However, the loading in-
creases more slowly than the average burnup. For example,
the design profile increased the critical loading by
and the new profile by ?10%. From the point of view of
uranium consumption, profiling is practical.
With uranium burnup, the maximum thermal loading,
gradually decreasing in magnitude, shifts toward the bulk
of the core. Thus the problem of correct profiling requires
time analysis of the burnup process at the edge of the core.
This problem has not yet been completely solved. The
redistribution of thermal loading in fuel elements at the
edge of the core for the design type of profiling is shown
in Fig. 9.
OPERATING CHARACTERISTICS
1. Reactor Fuel Elements
By design, the fuel elements of the reactor should operate at a thermal loading of (5-6)? 106 kcal/m2 ? h. The
achievement of that thermal loading and the dense packing of fuel element plates in the reactor core (fuel element
thickness 0.8 mm, water gap width 1.6 mm) made it possible to obtain high specific core loading (T1=1600-1700 kW
per liter, qmax= 4500 kW/liter). Although thermal engineering calculations revealed the possibility of operating
fuel elements at such high thermal loadings, there were widespread misgivings as to whether they could actually be
achieved. In addition, it was necessary to check the attainable U235 burnup.
Only experiment could furnish answers to these questions. Therefore, before bringing the reactor up to nomi-
nal power, a fuel element test was performed in channel 1 with a thermal loading of 6 ? 106 kcal/m2 ? h up to a maxi-
mum U235 burnup of 28%. In all, six fuel element plates were tested in this experiment. With reactor operation at
nominal power, a thermal loading of 6 ? 106 kcal/m2 .h was achieved simultaneously in a large number of fuel ele-
ments, and the maximum U235 burnup in the reactor core reached values of 35% and higher in individual fuelelements.
After a period of operation, two fuel element assemblies were removed from the core (because of suspected
loss of integrity) with a 2-3% U235 burnup which pointed to manufacturing deficiencies.
1221
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
In order to determine the possible increase in thermal loading on the fuel element plates, tests were made on
U235 burnup at thermal loadings of (10-11)? 106 kcal/m2 ? h. The maximum burnup achieved- in these experiments
was 30%. The results of the tests showed that the thermal loading at the surface of a fuel element be increased by
a factor of two over the design value, and only because of this may higher neutron fluxes be attained with the reactor.
Thus the operating experience with the reactor and the experiments that have been performed have indicated
the capabilities of fuel element plates at high thermal loading in an intermediate reactor.
2. Control and Protection Systems
After the installation of additional rods during startup and adjustment operation, the control and protection
system of the reactor proved to be capable of assuring the required excess reactivity consideration of the overcharg-
ing with fuel element assemblies to take care of U235 burnup. During reactor operation, the protective system as-
sured safety; the number of spurious responses was small, and there were no serious failures in the electronic equip-
ment or in the electrical circuits of the protective, power measuring, and automatic control systems. Preventive
inspection and regular maintenance assured continuing operability of the systems.
The sealed drives of the compensating rods, located on the lid of the reactor, never got out of order during
the entire operating period after the introduction of several improvements in the startup period. During operation,
a minimum of preventive measures was taken including periodic measurements of the insulation resistance of the
windings and a check on the adjustment of the control units.
Hydraulic drives were installed for the four safety rods located in the neutron trap. They operated well and,
during operation, required only periodic adjustment of the indicator system.
3. Some Information About the Stability of Specific Parts of the Reactor
The stability of materials becomes an acute problem in a reactor with high neutron flux. Therefore the con-
struction of such a reactor must provide for maximum interchangeability of parts.
In working out the plans for the SM-2, the question of the selection of reflector material was discussed many
times and at great length. However, the experimental data on irradiation with high total neutron fluxes which was
needed for the final decision was lacking. It was decided to use beryllium oxide as the first type of reflector ma-
terial. During operation of the reactor, observations were made on samples taken from the removable assemblies
and from the shafts of the compensating rods for the purpose of checking on the condition of the beryllium oxide.
The samples removed had been irradiated by a fast neutron flux (E 1 MeV); the total flux seen by the samples
was 1020_1022 nicm2.
The data indicated that a total flux of 1022 n/cm2 was a maximum for beryllium oxide.
At the present time, because of this, the removable beryllium oxide assemblies have been replaced by me-
tallic beryllium assemblies. The effect of the fast neutron flux on a beryllium oxide layer in the fixed reflector
is less, and the construction of the reflector provides for measures preventing the breakdown of the beryllium oxide
blocks.
Another very-important part of the reactor, which operates under difficult conditions, is the lining of the hori-
zontal channels passing through the lateral reflector at the edge of the core. The lining header is irradiated by a
fast neutron flux of 5 ?1014 nicm2 ? sec and by y radiation from the core. In addition, it is subject to an external
pressure of 50 kg/cm2. Lengthy operation of channel lining leads to considerable radiation damage of the material
in the walls. The latter can be held to acceptable values by replacing the channel lining from time to time. The
capability for replacement is provided for in the construction.
4. Operation of Technical Equipment
During the reactor operating period, all the basic equipment in the cooling loops and in the auxiliary systems
operated normally. Pumps in the primary and secondary cooling loops, heat exchangers, and a cooling tower insured
removal of reactor heat at any time of the year. The technical equipment of the experimental cooling loops also
operated flawlessly.
CONCLUSION
Operating experience with the SM-2 reactor verified the correctness of the basic physical and engineering
ideas incorporated in the design;
1222
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
1. A thermal neutron flux in the trap of about 2.5.1015 n/cm2 ? sec was achieved at a power of 50 MW.
2. The flux of fast neutrons in the core with energies above 1 MeV exceeded 1015 n/cm2?sec.
3. A maximum thermal loading in the core of 4.5 ? 103 kW/liter was achieved with a coefficient of varia-
bility in the core equal to ?3.
4. The fuel elements proved completely efficient at a record thermal loading of 6.106 kcal/m2 ? h, and, as
experiment has shown, this loading can be doubled. Furthermore, the attainable burnup exceeds 35%.
5. The main engineering units of the reactor?the cooling system, the recharging system, the control and
protection system, etc.?proved to be completely effective.
At the present time, plans are being developed for an increase in reactor power. The reactor vessel, the
main loop, the auxiliary systems and facilities will remain unchanged. The reactor power is being increased from
50 MW to 100 MW. The core will be built with fuel elements manufactured in accordance with previously devel-
oped and tested technology. The height of the core is being increased to 40-50 cm. The beryllium oxide in the
reflector will be replaced by metallic beryllium, the reflector being made up of individual, interchangeable blocks.
Water-filled cavities will provide for an increase in thermal neutron flux in the immediate vicinity of the horizon-
tal channels.
The increase in power of the SM-2 reactor permits:
1) an increase in thermal neutron flux in the trap to (5-8)? 1015 n/cm2. sec (depending on the number of
samples which are irradiated in the core), and an increase in fast neutron flux (energies greater than 1 MeV to
2 ? 1015 n/cm2 ? sec;
2) sample irradiation in the "hard" spectrum of the core;
3) an increase in average U235 burnup.
LITERATURE CITED
1. S. M. Feinberg et al., In Reports of the Second International Conference on the Peaceful Use of Atomic
Energy [in Russian], Dokl. soy. uchenykh, Atomizdat, 2, Moscow (1959), p. 334.
2. S. M. Feinberg et al., Atomnaya energiya, 8, 493 (1960).
3. Ya. V. Shevelev and V. K. Saul'ev, Atomnaya energiya, 14, 200 (1963).
4. N. I. Laletin, Atomnaya energiya, 14, 458 (1963).
1223
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
THE PGR PULSED GRAPHITE REACTOR*
I. V. Kurchatov,f S. M. Feinberg, N. A. Dollezhal',
P. I. Aleshchenkov, F. S. Drozdov, I. Ya. Emel'yanov,
A. D. Zhirnov, M. A. Kazachenko, G. D. Knyazeva,
F. V. Kondrat'ev, V. D. Lavrenikov, N. G. Morgunov,
B. V. Petunin, V. P. Smirnov, V. M. Talyzin,
A. G. Filippov, I. L. Chikhladze, P. M. Chulkov,
and Ya. V. Shevelev
Translated from Atomnaya fnergiya, Vol. 17, No. 6,
pp. 463-474, December, 1964
INTRODUCTION
For investigating processes taking place in various substances irradiated by neutrons and y rays a variety of
demands may be imposed on the radiation source. For example, it is possible that intense irradiation over a short
period may be required. Reactors with a small neutron lifetime are used for this, for example of the Godiva type.
In other experiments it may be necessary to maintain a high neutron flux for a longer period. Pulsed reactors with
high integrated neutron fluxes are necessary here. The pulsed graphite reactor (PGR) belongs to this type of reactor,
whose construction was reported at the Second International Conference on the Peaceful Uses of Atomic Energy by
S. M. Feinberg [1].
The integral neutron density per pulse, if no heat is given out, is calculated by the formula
dt = vf 1-0
n
f V 0
where i is the energy which is available for absorption per 1 mole of moderator; E is the absorption cross section
for neutrons with a velocity v by a mole of moderator, Ef is the energy release accompanying fission; e is the ab-
sorption probability of a neutron in the fuel (uranium); qv/ is the probability that a neutron absorbed in the fuel
will cause fission. The table shows that graphite is the best material for a pulsed reactor with a large integrated
flux per pulse, since it has a favorable ratio of the energy received, i, to the neutron absorption cross section.
For (1-0)/0 =1.0 (a very large reactor with a charge of ?100 kg uranium) the integrated thermal neutron
flux may be of order f nTidt P.12.2 ? 1018 cm-2 (the average neutron velocity 7= f vdi/f di is 5.16 ? 105 cm/sec). In
order to reduce the critical loading of the reactor it is necessary to increase the concentration of uranium in the
graphite. Thus, for (1?e)1e = 0.07, the charge is 7 kg but the integrated flux is reduced by a factor of 14. If the
average temperature of the graphite is reduced to 1500?C (assuming nonuniformity of heating), the average inte-
grated flux throughout the reactor still falls by a factor of 2.5, i.e., to a value of 0.6 ? 1017 cm-2. These are pre-
cisely the characteristic parameters of the PGR reactor; the reactor for producing short-duration but extremely
intense beams of neutrons and y rays for irradiating samples. In constructing the reactor there arose problems in
the study of its dynamics as a result of large reactivities and also in controlling the neutron flux according to a
specified time law. A study of the behavior of the reactor structure with a core of graphite blocks and homogene-
ously-distributed nuclear fuel at high temperatures is of considerable interest and so also is an investigation of the
safety of homogeneous uranium-graphite reactors as a result of the rapid injection of large reactivities.
The reactors has already operated without trouble for several years. The scientific supervisors of the project
were I. V. Kurchatov and S. M. Feinberg. N. A. Dollezhar and I. Ya. Emel'yanov supervised the engineering part
*Report No. 322a, presented by the USSR at the Third International Conference on the Peaceful Uses of Atomic
Energy, Geneva, 1964.
Deceased.
1224
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Cornparison of Moderator-Diluents for Uranium According
to the Magnitude of the Attainable Integrated Neutron Density
per Pulse
Moder-
ator
Permissible
temperaturee
T, ?C
J/mole
cm2/mole
f ndt for
= 0.5,
cm-3 ? sec
C
2700 (stability
lost)
6 ? 104
2.4 ? 10-3
4.2- 1012
Be?
1300 (stability
lost)
5.9 ? 104
6.0 ? 10-3
1.0 ? 1012
D20
300 (uranium
separates out
in residues
2.7 ? 104
6.0 ? 10-4
7.6 ? 1012
of the project. On the basis of the planning and adjustment of the control equipment, investigations into the dy-
namics of pulsed reactors were initiated and were carried out by Ya. V. Shevelev, who was also responsible for de-
veloping the methods of assessing the PGR reactor.
The principal physical characteristics of the PGR reactor are presented below:
Principal Physical Characteristics of PGR
Core Graphite, impregnated with uranium
(90% U235); dimensions 1400 X 1400
X 1400 mm; moving section 800 x
800 X 1400 mm; size of block 100 X
100 x 200 mm
Reflector Graphite; thickness everywhere not
less than 50 cm; dimensions of
stack 2400 x 2400 X 4200 mm
Nuclear ratio U/C 1:10,000
U235 charge, kg 7.46
Excess reactivity (22?2)? 10-2
Effective delayed neutron fraction . 0.00685
Prompt neutron lifetime, sec (1.07?0.03)? 10-3
Integral neutron flux with complete
withdrawal of rods, cm-2 1.1-1017
Maximum possible neutron flux in
the burst regime, cm-2- sec-1 . . 1 .1018
Operating Regime
The PGR reactor is designed for operating in two regimes: the self-quenching burst regime and the controlled
regime with a duration of several seconds or more. In order to achieve the regime of the first type, a reactivity is
injected into the reactor which exceeds the delayed neutron fraction. The rate of reactivity injection must be such
that after completion of motion of the trigger element several periods of excursion should elapse before the power
attains the vital magnitude and leads to heating up of the core. If this condition is observed, then it can be assumed
that the reactivity is being injected instantaneously and stepwise. The reactivity step r0=ke-1/Beffke determines
the initial period of the excursion, the amplitude and shape of the burst. The burst is self-quenching as a conse-
quence of the core heating up, since the reactor has a negative temperature coefficient of reactivity. As a result
of this the reactor is found to be subcritical at a value close to 1.0. The precise final state of the reactor is described
by the equation
15,x,
r (19) c115 = 0,
1225
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
co
where r(9) is the relationship between the reactivity and the accumulated heat 11= Qdt, ?00,, Qdt,
0
Q is the power. A considerable fraction Of the heating occurs in the slow stage of the process when the reactivity is
negative and the power is small. Forced quenching of the burst by rods reduces additional useless heating up of the
reactor.
The controlled regime also begins with a self-quenched burst, but at the instant of its peak the units which
compensate the drop in reactivity (the compensating rods) commence to move. The movement of the rods is pro-
grammed so that in the operative section of the pulse 0 t tp, the power is governed by a specified law Q=Qp(t)
(for example, it has remained constant). In future the following nomenclature will be used for analyzing the con-
trolled operating regime: Xi are the decay constants of the nuclei?of the sources of delayed neutrons; l'w=1/Baf
is the prompt neutron lifetime related to a eff: Y(t)= atlae' is the reactivity coefficient with respect to the accumu-
lated heat; Arp= I f y 01 is the magnitude of the reactivity compensated in the operative section;
A rp
YP t=tp
i=0
is the aVerge value of)' in the operative section, a (t) =- Ati,t1L-? ?1* , ----- T=- = X/*.
yp t p
The actual reactor characteristics of PGR are used AS an illustration,
The change of reactivity rp in the constant power operating regime, Q =cOnst, is described approximately
by the function
(t) = 1.? rp (t) = 1-3L (1 ?e?kit).
This function varies from zero for t=0 to unity for large values oft. For t < Ocr= 62/(a-8) Le 0.064; a(0)= ai
7sec/tp] the reactor is unstable (the formula is correct when a 3 8). But at the instant of the burst peak there
is, in fact, already a certain number of delayed neutrons; ;4=28/(4-1). Sirice ipi> :per, the reactor is stable right
from the beginning, i.e., perturbations lead ta finite deviations of the power from Q. However, these deviations
at small values of t (large values of a) bear the nature of weakly-dainped oscillations which are continued so long
as ip< 211.a. For >241 an aperiodic regime is obtained, For prolonged regimes the aperiodicity begins imniedi-
ately, TO an accuracy of up to 10%
69anax Or
QP
where (Sr is the stepwise change of reactivity. For identical values of 6r= 0.1 at the beginning and end of the oper-
ative section of the regime with a duration of 30 sec we obtain .5Qtnax/Qp =0.18 and (SQmax/Qp =0.12, respectively.
The product Qtp in an uncoOled reactor is limited by the maximum permissible temperature and therefore
any excessive heating is harmful: it leads to a reduction of tp at the time of emergence into the operative section
and on completion at the time when the power falls to zero.
The loss of operating time, associated with heating as a result of arrival at the operating section is equal to
oti *rmax
tp Arp
where &comp is the reactivity compensated at the instant of commencement of movement of the rods
P-ito -1); his the initial value of y,
For a regime with a duration of 5 sec (re =3.5) Sti/tp = 0.057. The loss of operating time at the end of the
process is considerable if forced quenching is not effected:
(Orcomp
1226
otf .ityp 2
tp V yf Arp%eff tp
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
1 I
I
.
IU
.111
.1
ME"
111
_!

? 11111111111111111ll
0+500
IIIIIIH
0
/
IP&
co'
e e ow 0.3.
111
Ellt ?
Or
[k
03500
02+00
_
VP
Fig. 1. Vertical section through the PGR reactor.
?o
1227
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
where Xeff is the effective decay constant of the delayed neutron sources, combined into a single group. For
tp= 5 sec and X eff X, Stf =0.35tp. Numerical calculation gives
Stf 0.3tp.
Forced quenching, if it occurs after a time which is small in comparison with 0.3tp, can be considered as an
instantaneous reduction of reactivity. It reduces very sharply the additional heating. If it be assumed that subcrit-
icality of the hot reactor with the rods inserted is equal to Arp, then as a result of the instantaneous reinsertion of
the shim rods
otf Teff Pi 1?e P
tp ? Arptp ' teff Xi
In a regime with a duration of 5 sec we obtain 64 = 0.03tp. As a result of reinsertion of the rods after a time
of order tp, numerical calculation gives 64 = 0.015tp. In fact, quenching by the rods reduces significantly the re-
activity and Stf 0.1tp.
Reactor Construction
The PGR stack consists of a number of graphite columns. Its dimensions are 2400 X 2400 X 4500 mm and it is
enclosed in a cylindrical steel hull with a diameter of 3100 mm and a height of 4500 mm (Fig. 1). The columns of
the core are constructed of uranium-impregnated graphite blocks with dimensions 100 X 100 x 150 mm. The blocks
are suspended on flexible graphite (cold) rods. The blocks, located one above the other, are not in contact with
each other which allows them freely to expand and prevents their jumping as a result of rapid heating. The columns
are mounted on the metal bearings of the lower reactor plates; there are gaps of 4 mm between columns. The upper
and lower sections of the columns consist of blocks which are not uranium-impregnated and which form the top and
bottom sections of the reflector; there is also thermal insulation between the hot core and the bearing structures.
The central sections of the columns comprise the core with dimensions of 1400 x 1400 x 1400 mm. The side re-
flector is constructed of graphite blocks, 200 x 200 x 600 mm. The central columns are mounted on fixed metal
stages and they form the fixed part of the core with a cross section of 800 X 800 mm. The stage is mounted on a
vertical hollow steel column which is connected with the bottom of the casing by a rubber bellows. When the mov-
ing section of the core is lowered, it enters the cavity of the lower reflector and is then surrounded by columns con-
taining boron absorber. The upper ends of the moving section columns are fastened to each other by graphite re-
tainers. The corner columns of the moving section and the adjacent columns of the fixed section consist of blocks
which are not uranium-impregnated. The gap between the moving and the fixed sections is 8 mm.
Above the moving section of the stack is located an additional suspended reflector formed by 100x100x 500 mm
blocks and suspended by metal rods which are welded to the cap of the casing. In the reactor stack there is a ver-
tical central experimental channel with a cross section of 200 x 200 mm. Nineteen channels with a diameter of
15 and 32 mm in the core and in the reflector serve for distributing thermocouples and also for a remotely-con-
trolled neutron source. Thirteen cold columns of the fixed section of the core have channels with a diameter of
65 mm for inserting rods. The operative section of the rods is constructed of five flexibly-coupled graphite tubes
with diameter 55 mm, filled with tablets of a graphite and gadolinium oxide mixture. The diameter of the ab-
sorbing section of a rod is 40 mm, with a gadolinium content of 0.02 g/cm3.
Thus, in the core of PGR there are no metallic structures and therefore the reactor heating is limited only by
the thermal stability of the graphite.
Because of the high design temperature of the core, particular attention has been paid to the uniformity of
distribution of the uranium in the graphite and to its identical content in the blocks. A special impregnation method
has been developed for introducing the uranium into the graphite, which is distinguished by its simplicity in techni-
cal equipment. The impregnation was carried out in vacuo. The method of depositing the uranium within the
blocks permitted simple drying operations at 105?C and subsequent calcining in air at a temperature of 400?C to be
carried out without danger of redistribution of the uranium in the blocks. The impregnated blocks were divided into
groups: a solution of specified concentration was used according to the graphite density. The average density of
the core graphite was 1.71 g/cm3 and 1.60 g/cm3 for the reflector. The concentration of uranium, enriched to 9010
in the isotope U235, was 3 g/kg in the graphite. The average ratio of the number of uranium nuclei to the number
of carnon nuclei per unit volume of the core is 1: 10,000; the average density of the core is 1.43 g/cm3. Uniformity
of distribution of the uranium in the graphite was controlled by the transmission of the block to a narrow beam of
1228
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
cold neutrons at several points and in two perpendicular directions. The uranium content of the blocks was deter-
mined by the count of fast neutrons from the block located in the thermal column of a reactor. In assembling the
core, blocks with a high uranium concentration were arranged in the less hot sections?at the periphery.
All metallic structures of the reactor are shielded from the effect of high temperatures by thermal screens.
The outer side surface and the upper part of the reactor vessel are "washed" with water. Vertical tubes pass through
the water in which counters and ionization chambers can be arranged. The lateral shield of concrete is lined
inside with metal. A layer of water above the cover of the counter (about 1 mm thick) and a steel dismountable
metal structure with a total thickness of about 400 mm (the stage of the biological shield) form the upper shield.
On the staging are arranged the servos for the reactor control system. Vertical steel tubes for insertion of control
rods are installed from the cover and the bottom of the casing and corresponding with channels in the graphite stack.
The upper tubes pass through the stage of the biological shield and terminate in flanges for mounting the control
rod motors in vacuum-tight containers. The lower tubes lead into a bunker situated below the reactor hull. The
design of the reactor provides for substitution of eight of the graphite rods by light rods of thin stainless steel and
cadmium. In this case, in place of the electric motors, membrane pneumatic equipment is installed which enables
the rods to be removed from the core in 0.05 sec; the bunker serves as a storage site for the spent rods.
The reactor vessel and the bunker connected to it by the tubes are filled with helium; the ampoule in which
the samples undergoing test are located, and which is cooled by water, is located in the central experimental chan-
nel. The PGR reactor has no special cooling system for the core. Its technological circuit includes in its gas-tight
loop, three water cooling loops (the assembly hull, the stage of the moving section of the stack and the ampoule in
the central experimental channel) and other auxiliary circuits. The ampoule cooling circuit consists of a heat ex-
changer, three vortex pumps and two centrifugal-vortex pumps operating from independent power supplies. The
centrifugal-vortex pumps are included for suppressing the gas phase of radiolysis products. In the circuit at the exit
from the ampoule is installed a shuttle valve, which directs the water into a settling tank during the time of oper-
ation of the reactor and into the heat exchanger after shutdown. Makeup of the circuit for water-cooling the am-
poule is effected through a system of filters. In order to remove corrosion products, a loop is installed in the heat
exchanger bypass for chemical purification of the water.
There is a special remotely-controlled device in the PGR reactor, designed for the withdrawal and transporta-
tion of the experimental samples into the hot laboratory.
Special Assembly Features of the Reactor Stack
Particular attention has been paid, in assembling the PGR reactor stack, to the question of operating safety,
as a consequence of the large planned excess reactivity of ?2010. The complex core configuration and the neutron
absorbers inserted into it has made unreliable the use of the normal methods of controlling the subcriticality of the
system (plotting the inverse count curves). The method of "shooting out" the neutron source by means of a pneu-
matic device has been employed. The high sensitivity of the equipment has enabled the system condition to be
reliably controlled in the process of assembly of the core. This same method has been used for estimating the com-
pensating efficiency of the rods, since measurements starting from ke = 0.95 are sufficiently accurate.
Assembly of the actual stack was preceded by assembly of a simulated stack which did not contain uranium.
The simulated stack was assembled in accordance with a program compiled for the assembly of the real stack and
it was found to be very useful for the subsequent assembly of the reactor. The assembly techniques are worked out
and the control and shielding systems were checked. Not least was the role played in the instruction of personnel,
since the reactor design is not very simple; the core alone contains 2384 items. The mock-up was dismantled ex-
cept for the lower reflector. The moving stage was withdrawn into the uppermost position and the moving section
of the core was mounted to it. Located in it was the central rod?with a steel sleeve of 100 mm diameter, lined
inside with cadmium and filled with paraffin for intensifying the compensating efficiency. The stage was then
lowered to the lowermost position, the moving section of the core was positioned within the reflector and a metal
dismountable structure?identical in configuration to the moving section of the core and lined inside with cadmium
?was installed in its place. It was designed both for reactivity compensation and for preventing collapse of the
columns of the fixed section of the core during assembly. First of all the columns fabricated from graphite which
was riot impregnated with uranium and with channels for inserting rods were assembled and cadmium rods in alu-
minum tubes with a diameter of 40 mm were inserted. Three rods, together with the central rod, were used as rods
for the scram system. After assembly of the fixed section of the core and of the upper reflector, the metal structure
1229
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
was removed. During the period of physical startup the central blocks of the upper reflector were suspended by a
vertically moving platform. On conclusion of the physical measurements the reactor vessel was sealed off by the
cover with a section of the upper reflector suspended by it. The temporary cadmium rods were subsequently sub-
stituted by the standard graphite rods. The operation of lowering the cover with the suspended reflector had been
previously rehearsed on the mockup stack. The principal possible emergency situations were calculated beforehand
on a computer using a two-dimensional program.
Physical Characteristics of the PGR Reactor
Considerable difficulties were encountered in the physical calculations, associated with the complex configu-
ration of the reactor and its inhomogeneous composition. In the core there are graphite columns which are not im-
pregnated with uranium, cold rods connecting the individual blocks, gaps between the columns, the central experi-
mental channel cavity, channels for rods and thermocouples. In addition, as a result of lowering the moving section
of the core a large-sized cavity is formed and the effect of the lower reflector with the boron absorber is manifested.
Despite a number of simplifying assumptions?rounding off the transverse reactor cross section, substitution of the
symmetrically arranged rods by an equivalent circular zone, the use of a diffusion equation in the central cavity
?the results of the calculations are found to be in satisfactory agreement with the experimental results. The calcu-
lation was carried out on a computer mainly in two-group approximation with the introduction of a refining cor-
rection which takes into account the fast neutron leakage. Some calculations were checked in nine-group approxi-
mation. The effect of slots, channels for the rods and thermocouples [2], and nonuniformities in the fuel distribu-
tion (of the cold columns) was estimated by perturbation theory.
The calculated values of (Ke-1)/Ke are given below as a function of the average core temperature for the
reactor without rods:
T, ?K
293.7
500
1000
1500
2000
2500
(Ke-1)/Ke
0.197
0.152
0.80
0.025
?0.022
?0.066
The magnitude of the excess reactivity is one of the most important characteristics of the PGR reactor, since it de-
termines the integrated flux per pulse. The calculated value of the excess reactivity is (Ke-1)/Ke ealc. = 0.19} +0.016
?0.018
and agrees satisfactorily with the experimental value of (Ke-1)/Keexp.
=0.22?0.02 obtained by the poisoning
?
method. Forty absorbers were inserted in the core, arranged between the columns of the stack in the form of a
regular lattice with a pitch of 20 cm. The absorbers were cadmium strips, sealed on both sides with aluminum
plates. Their compensating efficiency was chosen such that the reactor without rods and with raised moving sec-
tons of the stack became critical. For this, holes were drilled in the plates. Then the compensating efficiency of
each absorber was measured by means of two rods, previously calibrated by the excursion method. The total com-
pensating efficiency of all the absorbers was reduced by a factor of 1.08 (0.08 is the correction to the poisoning
heterogeniety, calculated to an accuracy of 10%). The value obtained is the excess reactivity. The error in the
experimental value of (Ke-1)/Ke is determined mainly by an inaccurate knowledge of the delayed neutron frac-
tion [3].
As a result of lowering the mobile section of the stack and with the rods removed completely from the core
and also with the upper reflector removed, the reactor was critical at a temperature 20?C. After inserting the ther-
mocouples and replacing the temporary cadmium rods by the standard graphite rods, the ends of which are located
at the boundary of the core and the upper reflector, the reactor under these conditions was subcritical at the value
I r I
2. When the rods are inserted in the core the critical position is obtained by raising the mobile section of the
stack to a level distant 530 mm from the upper position; if the mobile section of the stack is raised to the upper
position then the reactor is supercritical at a value of r=4.5.
The prompt neutron lifetime was calculated by the change of the effective multiplication factor Ke for an
ake
imaginary insertion of an absorber whose cross section varies according to a 1/v law, viz / ? _1
v a (neff)
value of /1=0.190 sec was obtained by varying (SEeff. The calculation was carried out for two positions of the
mobile section of the stack and it was established that Z depends weakly on its displacement. The prompt neutron
lifetime was determined experimentally for fast reactor excursions by the period of the excursion and by the known
initial reactivity generated by the movement of the calibrated rod. In two experiments (4=1.425 and ro= 2.05),
values were obtained of / =0.162 sec and /*= 0.153 sec. The mean value of /* is 0.158 sec.
1230
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
25
?0
1,5"
Declassified and Approved
For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
In order to determine the neutron flux, the peripheral
thermocouple channel of the core was used, the closest to the
SO mobile section?channel b-8?and equipped with a lock for the
1
2
10-
5
0
300 1000 1500 2000 2500
Fig. 2. Functions z(T), Y(T) and
(am
T ?
' cri0 C
1) Z= _ ; 2) Y=-- !0
(3. dT coTo
l'o i T (giv)
T0
C dT
3)15? coTo "
To
insertion and removal of samples without disturbing the vacuum
50 tightness of the equipment. It was assumed that the neutrons
which are incident on the chamber are fast neutrons which have
40 been slowed down in the water around the chamber, and hence
there should be a relationship between the chamber current and
30 the release of heat in the peripheral sections of the core, whence
only fast neutrons could reach the chamber. This relationship
20 is slightly sensitive to the condition of the reactor (to the posi-
tion of the rods and to the temperature) and it was determined
8 so that in future the neutron flux could be assessed firstly on
the basis of the chamber currents and secondly by the relation-
ship between the flux at a specified position and the flux in
K channel b-8.
cc 0 dr T per wire were activated in channel b-8. The ratio between
o ;
Gold foils (in cadmium and without cadmium) and a cop-
the activity of the wire and the thermal neutron integral den-
sity f ndt was determined by measuring the absolute activity
of the gold foils by the 8-), coincidence method. The cadmium
ratio for the gold was found to be 2.7. By establishing the con-
nection between the activation and the integrated flux on the
one hand, and between the activation and the integrated neu-
could be found for conversion from integrated fluxes to f ndt:
Iron density on the other hand, the factor
n dt =E iohdt.
The relationship between the integral neutron density, integrated flux and heating is established in the following
manner. If N5 is the number of U235 nuclei per kg of the core, C(T) is the specific heat of 1 kg of the core at a
temperature T and Tif(T) is the mean fission cross section of U235 by neutrons in the stack at temperature T, then
C dT = E iN 5-6 fni; dt,
where Ef is the energy released by a single fission. Hence
n dt F?Y, Y? (Qfv)? C dT F= coT 0
(aft) CoTo ' E1N5 (afiT)0
n?vdt = ? Z , Z C C dT G= c0T2
cri COT
E fl)lbcr fo
The suffix "zero" defines some reference temperature and T1 and T2 are the initial and final temperatures. With
large heating (of order 1000?C) the chamber current is associated essentially not with the quantity n but with the
quantity nvof , i.e., with the heat release C (dT)/dt. Consequently, the relationship
-= H ichdt; = CcociTTo .
(1)
(2)
(3)
(4)
should be fulfilled. The coefficient H can be found from comparison of formulas (1), (2) and (3), bearing in mind
that formula (1) is valid when af-17= afovo:
H F '
Thus, if the initial temperature T1 and the chamber current (1 ichdt) are known, then T2 and f nVdt can be found
(Fig. 2). Measurement of the temperature with the thermocouples makes it possible to adjust the accuracy of de-
termining the integrated flux. Thus, during the time of the first heating up of the reactor, when the shim rods had
1231
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
ar/d8
12
10
9
2
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
traversed one-half of their path, the temperature in channel b-8 as
determined by f ichdt was equal to 820?C. The temperature accord-
ing to the thermocouple readings was almost identical with this value.
In the above-mentioned triggering the integrated flux in chan-
nel b-8 was 2 ? 1016 cm-2 and at the center of the core it was 4. 1016
cm-2. According to calculation the reactor, without rods and with
the mobile section of the stack withdrawn to the upper position, be-
comes critical at an average temperature of 1800?K, which corre-
sponds to a flux at the center of 1.1 ? 10" cm-2. If the rods are not
withdrawn from the upper reflector, then this value is reduced to
0.8.1017 cm-2.
1 2 3 9 5
8
Fig. 3. Thermal reactivity coefficient.
The neutron fluxes in channel b-8 at the instant of the self-
quenched burst maximum are given below as a function of the mag-
nitude of the initial reactivity jump ro or the initial excursion period Te:
ro
Te
nv
5.50
0.035
1.34.1016
4.46
0.045
0.84.1016
3.65
0.059
5.2 ? 1015
2.54
0.10
1.98 ? 1016
1.47
0.285
2.9 ? 1014
The maximum possible instantaneous value of the neutron flux in the burst regime was estimated on the basis
of the calculated dependence of the reactivity on the accumulated heat r(9) for the reactor without rods and with an
average temperature of heating up of 1600?K;
1,1
KG dZ
(n17.)max = (r ?1) dO = 1.1018 cm-2 ? sec-1.
Here 9?1 is the value of .9. for which r= 1; Kv is the coefficient of volume inhomogeneity.
The temperature dependence of the reactivity was found for small values of .9- during the first reactor trig-
gerings?bursts with heating up of the core. The reactivity was calculated as a function of time from the oscillo-
gram of the ionization chamber currents by means of a special program on the computer and, according to the in-
tegral of the chamber current it was
r = rod- lc, k [Sichdt12 .
The relationship between the reactivity and the accumulated heat was expressed in the form
r =Ro x05+ x219.2.
The coefficients xi= ?12.6 and x2=5.7 were determined from these relationships and the function Ar= 49-2
was plotted, describing the dependence of the reactivity on the temperature in channel b-8. The temperature coef-
ficient decreases according to the extent of heating up of the reactor; its initial value is dr/dT = ?0.042 deg-1.
In the controlled regime the reactivity coefficient with respect to temperature depends only on the displace-
ment of the shim rods. In order to derive it, the dependence of dr/dxrod on the position of the rod xrod was used.
The efficiency of the shim rods in the critical reactor was determined as a function of their position (with identical
extent of withdrawal of eleven rods) during cooling down of the reactor which had been heated to 1000?C (the tem-
perature of channel b-8). The measurement procedure is as follows. Having waited until the cooling reactor be-
comes slightly supercritical, one of the rods is moved a few centimeters upwards, the change of reactivity is deter-
mined and all the rods are lowered to a new level. The change of reactivity found is divided by the displacement
of the rod and multiplied by the number of rods. The temperature field during cooling down does not reproduce
exactly the temperature field of the heated reactor, nevertheless the relationship dr/dxrod= -r(
,xrod) obtained, was
used for calculating the reactivity coefficient by the parameter & for channel b-8. Oscillograms of prolonged
1232
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
...20 t, rex
Fig. 4. Triggering oscillogram, ? ) Ionization chamber current
[ich(t)]; ? ----) specified law of current variation [i(t)I.
regimes carried out with the regulator disconnected were used. The oscillogram of the current ich(t) and the known
coefficient H make it possible to determine the relationships a(t) and r(t), r(a). The relationship xrod(a) is
found by means of the xrod(t) oscillograms. Further, dr/da is determined from the relationship dt/da=(dr/dxrod)?
(dxrod/de)+
(dr/da). The graph of the relationship dr/a.= f (9) is shown in Fig, 3.
PGR Reactor Control
The control mechanism of the PG 11 is divided functionally into trigger, compensating and control. The trigger
element, which initiates the neutron burst, is the moving section of the stack. For triggering the reactor it is always
raised to its top position and therefore the magnitude of the initial reactivity jump ro is determined by the initial
temperature of the stack, by the position of the rods and by the absorption of the samples in the ampoule of the
central experimental channel. The movement of the eleven compensating (shim) rods is started at the instant of
the burst maximum, after operation of the threshold equipment whose detector is an ionization chamber. Synchro-
nous movement of the rods, by a law which is defined by the profile element, is ensured by a servo-system. An
automatic control rod corrects any errors.
The first triggerings of the reactor were undertaken with identical withdrawal of the shim rods and with the
automatic controls detached. The ionization chamber currents ich(t) were oscillographed and also the movement
of the rods xrod(t). At the coordinates xrod(0;
J(t), where J(T) = i? , triggerings of different duration were
represented by similar curves, since in any triggering after reaching the operating region the reactivity differs but
little from zero (by not more than unity). The relationship xrod= f (j) also represents, to a first approximation, the
law of motion of the shim rods in the regime ich(t)= const. A more precise law was obtained after working out the
results of several triggerings with a previously adjusted profile element. Because of the different extent of the
graphite temperature effect on the temperature of the neutron gas in the core and in the ampoule, the ratio of the
chamber current to the neutron density inside the ampoule does not remain constant. In order to adjust the profile
element to the regime in which the neutron density inside the ampoule should remain constant, special measure-
ments were carried out. In particular, a loop wire ?activator was positioned inside the ampoule which was pro-
vided with a device for exposing the wire by sections. The section of the wire irradiated for several seconds was
discharged, after each exposure, from the core and collected in a container which was shielded from thermal neu-
trons. By comparing the activation of the individual sections of the wire with the current integrals during the time
of irradiation, it was established that the chamber current should be reduced towards the end of the regime (Fig. 4),
The procedure worked out for the calculating allows a profile element to be produced for any operating re-
gime of the reactor. In particular, there is the interesting possibility of producing neutron pulses of different dura-
tion, which approximate in shape to rectangular. In this case, the operating time losses at the beginning and end
of the process will be minimum. A regime of this type can be divided into three stages: the initial stage, during
which time the power of the reactor reaches a specified value, the operating region (Q=Q ) and the final stage is
1233
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
1,0
0,5
0
A
1 \
t
I
I
I
I
I
Fig. 5
t sec
0,
0
/Af
I
I
/
k
1
1
/
2
e'-?,
%
\
/
/
`..7.1
__-
i
2 3
Fig. 6
5 t,sec
Fig. 5. Effect of errors in the reactivity jump 6r0 on the behavior of the power in the regime tp= 10 sec;
1) 64=-0.05; 2) Srt =0; 3) 64=0.05.
Fig. 6. Effect of errors in the instant of commencement of motion of the shim rods on the behavior of
the power in the regime tp=10 sec. 1) Stred = -0.05 sec; 2) 6trod = 0; 3) Otrod '0.05 sec.
the forced quenching. The most important is that of reaching the operating region, since initially the reactor has
the least stability. Successful arrival at the operating regime is determined mainly by the correct choice of the
value of the initial reactivity jump r0 and the instant of start of the movement of the shim rods. Errors in the re-
activity or in the response threshold of the switching system, Qthresh, of the shim rods give rise to considerable
power fluctuations at the beginning of the operating section of the regime.
The relative error in the maximum power of the burst depends on the error in r0 thus:
8Qmax =__ 215r0
Qmax r0-1
The error at the instant of commencement of motion of the shim rods otred leads to a reactivity perturbation:
orpert.. 8ttrOd1jAr p; (yi .117p) Ar p 50.
P
The 'error in r0, equal to 0.05 in a regime with a duration of 20 sec (4=1.78), changes the amplitude of the burst
by 13%; a shift of the instant of engaging the shim rods by 0.05 sec is equivalent to a reactivity perturbation at the
commencement of the operating region of 6r ,-:_10.13. Figures 5 and 6 illustrate the effect of errors in the initial re-
activity and in the instant of the initial motion of the rods on the behavior of the power. The curves are obtained
by a numerical method by means of special program on the electronic computer.
A number of experimental relationships are used in making a correct choice of r0 and 0
_thresh. The value of
the ionization chamber current (the unit of the servo-system of control by the rods) is determined at the beginning
of the operating region, ieh(ti), from specified values of tp and Q. The magnitude of the initial reactivity jump
is chosen from the condition ich max (1.0' ich(ti)-
In determining 0
-thresh account should be taken that the movement of the rods commences with a delay of
L\ trod after actuation of the threshold circuit. Consequently, it is necessary to know not only the amplitude but also
the shape of the ensuing burst. A value ofQthresh is determined for which initially the motion of the rods should
coincide with the burst peak.
If the position of the rods is fixed and there is no sample in the ampoule, then r0 as a result of raising the
moving section of the core is determined only by the temperature of the stack. The relationship between r0 and the
temperature (when it is uniform over the whole stack) can be represented, within the range 20-50?C, in the form of
1234
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
a straight line with a slope of dro/dT = ?0.034 deg-1. The temperature must be measured with an accuracy of a
fraction of a degree, otherwise the error in ro will be inadmissibly large. In the initial state, prior to triggering,
the shim rods will be located in the strictly fixed lower position and the automatic and manual control rods will
be in some intermediate position. By changing their position the magnitude of rc, is also changed. Reduction of
ro can be achieved also by a small preliminary heating of the core. The losses in integrated flux as a result of this
will be small, since at low temperature the specific heat of the graphite is small and the fission cross section is
large.
The compensating efficiency of samples installed in the ampoule is estimated by the position of the moving
section of the stack corresponding to the critical state. The most accurate value for the compensating efficiency
of the sample can be obtained by producing a burst which is quenched while the reactor is heating up (pseudoburst).
The riding up period and the reactivity are determined by the chamber current oscillogram. After the personnel
had acquired adequate practical experience, the error in attaining a specified value of ro in the majority of reactor
triggerings did not exceed values of 1.0= f 0.02. The power at the start of the operating region is also found to be
sensitive to the inaccuracy of adjustment of the profile element, but one and the same profile element is suitable
for different, but not too strongly different, values of tp. Figure 4 shows the form of an actual triggering with a
duration of 30 sec, carried out with the profile element for the regime t=20 sec.
The reactor has been found safe and convenient to operate as a means of producing high, although also short-
duration, neutron (and y ray) fluxes. In reactors of this type, with an enlarged core and a reduced uranium content,
fluxes of up to 2.1018 cm-2 can be attained.
The injection of reactivities which are many times greater than 13 is safe: as a consequence of the negative
reactivity temperature coefficient the reactor, as a result of this, reaches a certain limiting temperature. Control
has proved feasible to an accuracy of +310 by the behavior of the flux with time, by means of relatively simple
equipment.
The individual columns of blocks impregnated with uranium and connected to each other by graphite rods
are very flexible and they are unstable. However, the columns when connected together at the top form a system
which is stable on the whole, although it is a moveable and swaying system, and which easily withstands repeated
movements (up and down). No jamming has been observed, despite the contact with the surrounding stationary
stack. Also, no noticeable effect whatsoever of the stack column vibrations on the reactivity has been observed.
The graphite, fireproof rods with the absorbing gadolinium oxide tablets have proved to be completely efficient.
No significant volatilization of uranium and fission products from the stack has been observed as a result of
operating the reactor. There has been no appreciable contamination by fission products of the caps of the control
and safety rods directly connected with the core space.
? In conclusion, we take this opportunity of expressing our sincere appreciation to V. V. Goncharov for the
sponsorship and assistance which he has continuously rendered in the construction of the reactor, and to B. V. Kur-
chatov for direct participation in developing the technology for impregnating graphite with uranium. We also ex-
press our appreciation of the scientific workers, engineers and builders, whom we have not found it possible to men-
tion here, for having participated in the construction and commissioning of the PGR reactor.
LITERATURE CITED
1. S. M. Feinberg, Data from the Second International Conference of the United Nations Organization [in Russian],
10, Discussion of reports P/419 and P/1848.
2. Ya. V. Shevelev, Atomnaya tnergiya, 14, 4 (1963).
3. G. R. Keepin et al., Nucl. Energy, 6, No. 1/2 (1957).
1235
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
STATISTICAL REACTOR KINETICS EQUATIONS
(UDC 621.039.512)
A. B. Govorkov
Translated from Atomnaya nergiya, Vol. 17, No. 6,
pp. 474-479, December, 1964
Original article submitted December 13, 1963
In additipn to the kinetic equation for the flux of prompt neutrons, which determines their mean be-
havior in the reactor, we have the kinetic equations for the "production densities" of the neutron fluxes
which describe the statistical behavior of the prompt neutrons in the reactor. Taking into considera-
tion the space and energy distribution of the neutrons, a method is given for making an effective aver-
aging of the statistical behavior characteristics of the reactor. A discussion is given of the theory of
the Rossi a experiment.
In the elementary theory of statistical methods of determining reactor parameters, the behavior of neutrons
in a reactor is described by means of the one-group and the one-zone approximations [1]. However, there is some
interest in a more general treatment which includes the space-energy distribution of the neutrons. In the present
paper, a treatment of this type is made for the prompt neutrons, while the delayed neutrons in fission are taken ac-
count of formally in the external source. A similar problem has been treated in the papers of [7] by a somewhat
different method.
A neutron located in the reactor has the coordinate r, the energy E, and the direction of its velocity n = v/v.
In abbreviated form, the state of the neutrons (x E, n) is designated by the letter x, and the element dVdEdSZ is de-
signated by the symbol dx. The symbols x and x + h designate the finite volume of the reactor, and the finite en-
ergy intervals and the directions of the velocities.
The probability of having in the reactor, at a chosen instant of time, a given number of neutrons, distributed,
in addition, in a definite way over the points x, is designated by WN (x1, x2, xN; t)dxi, dx2, dxN. We
average the production densities of the neutron fluxes over these probabilities at several different points;
.F' (x' , x", . . . , x(s); t)
dx, . . . dxN[ E vis ] [ vis(xj_x")]
[
k=1
(j1) (h/=i, j, ? ? .)
Po
N=1
WN (Xi, X2, . . XN; t).
vh6 (a-h?x(8))]
(1)
The above densities have been given the name of production densities in the theory of random point proces'ses [2].
The factorial moments of the number of neutrons in the finite interval x, x + h are expressed in the following way;
xd-h
(xi ?
N (N ?1) . . . (N ? s 1) = dx, . . . dFxs s ' . 'x' 1) s
1.71 . Vs
X X
The production densities satisfy the equations; [see Eq. (3) on following page]
1236
(2)
Declassified and Approved For Release 2013/03/14 : CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
OF, . ? ?, xs;
... {?niviFs(x1, ..., xi, ..., xs; t)
at
5=1
F s (Xi, . Xi; Xs; t)?
dx:F (x1' ? ? x pc.c) (x: I x ?1
las, (xi) Ise (xi)
CO
dx FS (Xi X17 ? ? ? , Xi?I, Xi+1, ? ? , XS; t) [ V dx" dx(v)
lf (X) V!
V=I
x x" , x(v))1?Q (xi; t)
v.
X F , xi-1, xi+i, x; t)} -I-
v ... vs
11=2 ? s
F (x, xl, xj1_17 xs; t)
X s
CO
if (x)
x [ E v (v_I)
dx(Fl) . . . dx(v)p) (xl x1. . . . , x, x(11+0 , . . . , x(v))1 ,
s = 1, 2, . . .
(3)
Here, 1//a(x), 1//sc(x), 1//f(x) are the macroscopic cross-sections for capture, scattering, and fission respectively,
occurring when a neutron collides with a nucleus. The total macroscopic cross-section for all processes is lhasf(x).
The density of the relative probability that the neutron as a result of scattering at the point r will change its en-
ergy E and the direction of its velocity a to E' and n' is written in the form:
p(sc)(x I x') = p(sc) (r, E, n1E' , n') 6 (r ? r').
This function is normalized in the following way:
p(sc) (x x') = 1. (4)
The densities of the relative probabilities that as a result of fission of a nucleus produced at the point r by a neutron
with energy E and the direction of its velocity n, v secondary prompt neutrons will be formed with energies E', E",
E(v), and the directions of the velocities n', n", WO, may be written in the form:
a) (X I , . X(v)) = pt? (r, E, n E', . , E(v), neo) 6 (r? r') .. 6 (r ? r(v)).
These functions are normalized in the following way:
Edx' . . . dx(v) p(f) (x! x', . . . , x(v))=.1..
(5)
The density of the probability of emission in unit time by an external source of one neutron at the point r
with the energy E and the direction of its velocity n is Q(x, t) Q(r, E, n; t).
The function Fi(x, t) is the usual mean density of the flux of prompt neutrons, while Eq. (3) for s = 1 is the
usual kinetic equation. For s = 2, we obtain the kinetic equation for the "binary density of the fluxes," discussed
previously by the author as applied to the problem of the spread in pulse amplitudes in a pulsed reactor [3].
The Eqs. (3) may be solved successively, since the equation for each successive function includes expressions
depending on it itself and on the preceding functions. This "breakup" of the functions into following and preceding
is based on the assumption that the inherent liberation of energy in the reactor, which accompanies the nuclear fis-
sion processes, has no effect on the magnitude of the critical parameters of the reactor. Otherwise, Eqs. (3) are
1237
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
incorrect. For example, for pulsed fast neutron reactors of "Godiva" type, our treatment only holds in the stage of
the pulse where the liberation of energy in the chain fission reacting does not yet have any effect on the value ot
the critical parameters of the reactor. But the very start of the neutron chains in the individual pulses may fluctuate
greatly, although after the chains are sufficiently developed, the fluctuations stop [4].
The boundary conditions for the production densities are formulated in a way similar to the boundary condi-
tions for the uspal,neutron flux density. For a reactor bounded by a convex surface, and surrounded by a void, they
are of the form:
(ri, El, bn Ei, Int int
\..., 1'8, Es, 128; t) .0,
i = 1, s,
(6)
where ri bn designates a point on the boundary, and ni int is the direction of the velocity of the neutron inside the
system. The number of boundary conditions for each function is s.
For the majority of cases of practical interest, the solution of Eqs. (3) for the production densities may be as-
sumed in the form:
F (xi, . . . , xs; t) -= F8 (xi, x8) cps (t).
(7)
Then, Eqs. (3), using transformations analogous to the transformations of L. N. Usachev for the usual kinetic equa-
tion [5], may be written in the following way:
dcps (t) keff
(s) dt s k ff (Ps (t)
+ sq (s, t) Ts-1 (t) 4- ) (s) (t).
Here, keff is the multiplication factor for prompt neutrons,() is the binomial coefficient, and T(s) is the effective
lifetime of the neutrons in the reactor:
(8)
T (s) = d ? ? , dx3.1+ (xvii) vF-s8)
The worth of the prompt fission neutrons is:
Xil ? ? I S ? (9)
(s) = dx2. ? ? S dx, F+ (x2) ? ? F+ (xs) dx dx' F(x, x2, . ? ? , xs)
V2 ? ? ? Vs f (X)
X [ 7vVi- dX" ? ? ? dx(v)p1) (x I x', e, x(,)) F (x')} .
(10)
The quantity q(s, t) has the sense of the power of the external neutron source, including the worth of the neutrons:
1
q (s , (s) d F+ (x1) Q (x1, t)dx2. ? ? dxs F+ (x2)?? ? F+ (xs) p3
02 . ? vs - (..21 ? ? . 8). (11)
The coefficients y(s) which relate the different functions 95(0, are expressed in the following way:
? -1-? +
(a)(s) ?S dxg4.1 ? ? F (zp,+1) ? ? F (x3) ? dxs dx dxi? ? dxt,Fs?P.+1 (x, , ? ? 3)
vp.+1 ? ? ? vs 1 f (x)
1238
? jv ?I 1) ? ? ? dx(v)p(i) (x xi, . . x(3+1)
v =IL
x F+ (xi) ? ? ? F+ (x)}.
(12)
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
In all these expressions, the function F+(x) is a characteristic solution of the equation conjugated with the usual
kinetic equation with no external source for a reduction in neutron multiplication by the factor keff [53.
It is very inconvenient that the quantities T(s), q(s, t), and yu(s) depend on the subgcripts s of the filtictiori
cps(t), and can thus not be regarded as universal characteristics of t6 reattor kinetics. In Many cases, however, it
is possible to make use of the additional assumption that the production densities decompose into the product of the
densities of the usual neutron flux:
s) F (xi) F (x2) ? ? ? F (xs).
Then, we obtain the mean lifetime Of a neutron in the reactor:
The worth of the prompt fission neutrons is:
=
The power of the external source is:
and the coupling coefficients are:
T dx F+ (x) F (x) ;
vv!
dx" ? ? ? dx(v)p(vf) (x I x' x"
L
v?i
(1(0 dx F+ (x) Q (. ,
(13)
(14)
(15)
(16)
? dx dxi- ? ? S' dxp, f((xx)) [ 2 v (v_1) ? ? ? (v_p.+1)
v!
17-=
X dX(IA+1)? ? ? ? S' dx(v)p(j) (x xl, . , . , x, x(1+1) , . . . , x()) .1 F+ (x1) . F+ (17)
The set of Eqs. (8) determines the statistical behavior of the neutrons in the reactor.
We now determine the number of coincidences in counting neutrons with a detector in the time intervals
dt and dto, separated by the interval t?to (Rossi a experiment [61*). The probability of the event under discussion
is defined in the following way:
R (t, to) dt dto =- Z?? dxl? ? ? d, = dxj? ? ? dx'N 1 AT!N 0!
N=1 No
dx 8 (X) S doe' (x [ E 6 (x'? xi) [ E 6 (x
5,1
(i;, . . t I3C1, . . xj-1, Xj+1, . x,N0; to)
X w0 (xi, . XN0; t0)} dt dto?
Here, e(x) is the neutron counting efficiency at the point x (macroscopic absorption cross-section Of the neutron
counter, multiplied by its velocity): the function,
(18)
'It should be noted that independently of Rossi, a similar idea for an a experiment was proposed in the Soviet Union
by Ya. B. Zel' dovich and N. A. Dmitriev.
1239
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
WN, No-1 . . ; t Ix1, . . . , j_1, x?1, . . XN0; to)
is the probability of finding N neutrons in the reactor at the points xi, xN' at the instant of time t, if, at the
instant to after counting one neutron at the point xj, No - 1 neutrons remain at the points xl, xj_i, xj+i, xNo.
Multiplying this function by E 6 (x' ? xi), summing over the number of neutrons N and integrating over the
i=1
arguments xi, ..., xi,/ on the right-hand side of Eq. (18), give the density of the mean number of neutrons at the
No
point x' at the instant of time t for the condition that at the instant of time to, this density is 6 ? X')
? 6 (xj ? x'). The solution for this density may be found in the form of an expansion in the eigenfunctions of
the kinetic equation cDn(x), corresponding to the eigenvalues an. After substituting this expansion in the right-hand
side of Eq. (18), using the relation (1) in summing over the number of neutrons No, and integrating over their posi-
tions xl, xN0, we obtain the following expression for the coincidence frequency:
R (t, to) dt dt0=Sdx 8 (x) F (x) dto dx' 8 (x') F (x') dt
v'
dx dx" (x) [F (x, x") 4_ F (x) Q (x")]
s
ean (i-to) dx' 8 (x') (1:111 (x') dt
v' vv" any
dx (1)-7-F' (x) (Dm (x)
(19)
The functions F(x) and F(x, x") are the steady state solutions of the usual and binary kinetic Eqs. (3) for the reactor.
The first term on the right-hand side of Eq. (19) gives the usual random coincidences in counting neutrons in the
time intervals dt and dto, and the second term gives the coincidences due to the genetic relation between the neu-
trons in the chain reaction. If the period of time t - to is considerably greater than the time interval required to
set up the characteristic distribution in the reactor (t - to ? n = 2, ...), in the second term on the right-hand
side of Eq. (19) we can keep only the first term of the series in n, corresponding to the eigenvalue:
keff ?Uj_1
T keff ?
(20)
In order to simplify the preexponential multiplier in the resulting expression, we make an additional assumption.
We shall assume that quite good equality is maintained between the steady state solution of the kinetic equation
and its first eigenfunction, and that the binary density may be written as the product of two such solutions:
F (x, x")=1;p2F (x')F (x"),
(21)
where cp2 is determined by the steady state Eq. (8) for s = 2. After all the transformation, we obtain:
R (t, to) dt dto-= e2N2 dt dto e2N F2/2 x [ t?t? (
exp 1
(i/keff)-1 T keff dtdto,
where the mean number of neutrons in the reactor, from Eq. (2), is:
IV= dx F(x)
v
The averaged neutron counting efficiency is expressed in the following way:
dx (x) F (x)
dx F (x)
1240
(22)
(23)
(24)
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
The coefficient is;
X
co
v (v71)
v=2
'F (x)
dx_S ds' dx"
I f (X)
r2 T dx dx' F(x)
f)
"' ? ? ? , .dx(v)p(j) (x I x', x", x" , . . x(v)) F* (x') F+ (x")
" ? ? ?
dx(v)pV) (x x', x", . . . , x(v))1 F+ (x912 {1 I dx F v(x)
(25)
In conclusion, I express my gratitude to Yu. A. Romanov and N. A. Dmitriev, for discussion of the results of
the work.
LITERATURE CITED
1. F. Hofman, Scientific and Engineering Bases of Nuclear Power, Vol. 2 [Russian translation], Moscow, IL
(1950), p. 103.
2, M. S. Bartlett, Introduction to the Theory of Random Processes [Russian translation], Moscow, IL (1958).
3. A. B. Govorkov, "Atomnaya tnergiya," 13, 152 (1962).
4. G. Hansen, Nucl. Sci. and Engng, 8, 709 (1960),
5. L. N. Usachev, Reactor Construction and the Theory of Reactors, Paper presented by the Soviet delegation at
the Internation Conference on the Peaceful Uses of Atomic Energy (Geneva, 1955) [in Russian], Moscow,
Academy of Sciences Press, USSR (1955), p. 251.
6. J. Orndoff, Nucl. Sci. and Engng? 2, 450 (1957).
7. L. Pal, Suppl. Nuovo Cimento, 7, 25 (1958); L. Pal and H. Nemet, Pile Neutron Research in Physics, Vienna,
IAEA (1962), p. 491.
1241
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
CHANNEL EFFECTS IN FISSION OF EVEN-EVEN COMPOUND NUCLEI
(UDC 539.173)
L. N. Usachev, V. A. Pavlinchuk, and N. S. Rabotnov
Translated from Atomnaya Energiya, Vol. 17, No. 6,
pp. 479-485, December, 1964
Original article submitted December 12, 1963
An analysis is made of the experimental data on the fission of even-even compound nuclei in the
reactions (d, pf), (y, f), and (n, f) at the threshold. Calculations are made of the fission which is
based on the Bohr-Wheeler formula. All the experimentaldata given are in agreement with the
structure of the fission channels proposed by A. Bohr. It is found in this case that the fission chan-
nels with negative parity lie 0.6-0.8 MeV higher than the first channels having positive parity. For
the fission channels arranged in this way, the Bohr-Wheeler formula gives a quantitative descrip-
tion of the mean fission widths for the resonances of the (n, f) reaction in the nuclei U233, U235, and
Puml. To account for the large difference in the case of Pu239, it is necessary to assume that the
ground state of this nucleus has negative parity.
The experimental data are being analyzed for the purpose of finding what effect the spin and parity of the
fissioning nucleus have on the fission barrier. If the barrier may be assumed to be approximately parabolic, the
form of the barrier is completely determined by two quantitites, the height Ef and the parameter Ecurv, which
gives the curvature of the barrier at the saddle point Pl. The principal theoretical ideas as to the relation be-
tween the fission process and the quantum characteristics of the fissioning nucleus were developed in the papers
by N. Bohr, A. Bohr, Hill, and Wheeler [1-3], which presented a number of points of view used in the present pa-
per. These may be generalized in the following way:
1. At the saddle point, a considerable part of the excitation energy goes into potential energy of deforma-
tion, the nucleus "cools off," and the number of ways of exciting degrees of freedom not associated with fission
is greatly limited. There are accordingly not a large number of ways,?"fission channels"?by which the nucleus
can pass through the saddle point.
2. For even-even nuclei at the saddle point, there will be pairing effects, and the spectrum of the fission
channels will be similar to the spectrum of the low lying excited levels. Accordingly, on A. Bohr's hypothesis,
the fission channels are arranged in the following way in order of increase in the fission thresholds corresponding
to them (Fig. 1): 1) the rotation band of positive parity, based on the level 0+ (there is no 1+ level in it), 2) the
rotation band of negative parity, starting with the level 1-, the distance of which from the level 0+ we designate
by Ai (there is no 0- level in this band), and 3) at a distance of A2 kt 1 MeV above the lower of the fission chan-
nels (the channel 0+) there begin to be channels corresponding to single-frequency excitations, which, generally
speaking, may have any characteristics, including 0- and 1+.
3. For the mean fission width of the states with spin J and parity II, we have the Bohr-Wheeler formula:
P7, 11 JU = P (E(fi) curv, E),
21c
(1)
where V,11 is the mean distance between the levels of the compound nucleus, i is the number of the fission chan-
nels, and P is the penetrability of the barrier for the i-th channel, which is given by the expression [1]:
1242
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Fig. 1. Schematic representation of the struc-
ture of the fission channels corresponding to
A. Bohr's hypothesis.
T,
T2
Fig. 2, Schematic representation of the fission
curves in the (d, pf ) reaction, as found in [4].
Values of the Parameters E?), 42), E(1-) , and E(2) , Determined from the Data Found in Different Reactions, MeV
curv curv
1 Compound
nucleus
E(I) E(I)
I Clint
E(2)
t
E(2)
dry
Bn
T h232
U234
U236
13238
pulto
pu242
5,7 (V. f)
5,5 (d, pf)
6,0 (d, pf)
6,0 (v, f)
5,0 (d, pf)
-6,0 (n, f)
0,41 (y, f)
0,45 (d, pf)
0,58 (d, pf)
0,68 (v, f)
0,31 (d, pf)
_
6,4 (y, f)
6,2 (d, Pt)
6,4 (y, f)
6,6 (n, I)
6,6 (v, f)
5,8 (d, pf)
-
0,68 (v, f)
0,45 (d, pf)
_
0,9 (y, f)
0,38 (d, pf)
-
6,4
6,8
6,4
6,0
6,4
6,4
P (Er, E(c2rv, , E) =
(E(;)-E) ?
1 +exp
E(i)
curv
Equation (1) is used to find the characteristics of the fission channels.
The (d, pf ) Reaction
In [4], using the (d, pf) reaction, curves were obtained for the fissionability of the compound nuclei U234,
U236, and Pu24? below the neutron binding energy in the corresponding nucleus. The curves have the characteristic
form shown in Fig. 2, Below the neutron emission threshold, the only process competing with fission is emission of
y -quanta, so that the fissionability measured in [4] is expressed by the formula:
1
(2)
J, n \ J, II d, pf E , i.
- a
F --.- \ / . p ad?p E ,J , I pI '
rid, p I- I', n+ r r, n /
J, n f V 'd,
where oli$ pri is the mean cross-section for the formation of a compound nucleus with spin J and partity ii in the
d, p reaction, and the partial mean fissionability is obtained by averaging over the distribution of fission widths
and is equal to:
J
/ r _
S (a).
1+a
t v
(3)
(4)
1243
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Here, a =and S(a) is the "fluctuation function" (see below). It follows from Eqs. (1)-(3) that the fission-
ability curve will have the characteristic step form observed in the experiment, for one of the following two cases:
A. If f3c ? f- , even for P = 1 (i. e., on the plateau), the quantity ff may be neglected in comparison with
F in the denominator. Since r is only very slightly dependent on the energy, the fissionability will be simply
proportional to the fission width, and hence to the penetrability. On the plateau, the fissionability takes the value:
plateau j
n ff plateau J, H
fi Od, p
(Tcl, p
? _J: 11 _plateau J, 1-1
rf - rf P.
B. If, at some energy, the mean fission width is comparable with the radiation width, and continues to in-
crease more rapidly, then, for ri ? ry, the fissionability at the plateau is given by the ratio:
Fplateau6'cri:
ad, p
(5)
(6)
If, for the same energy, there are barriers for several combinations of the moment and the parity, Eqs. (5)
and (6) must be summed over all such combinations.
Assumption A is used in [4], from which it follows that the threshold value of the energy coincides with T
(see Fig. 2),?the point on the energy axis where F = (l12)Fplateau. The parameter Ecurv = (r/2)5E = ir/25 [here S
is the slope of the curve F'(E) = F(E)/Fplateau at the point where F' = 1/2].
For this interpretation to be correct, it is necessary for the value of ff , with a penetrability equal to unity,
to be several times less than r for the energy corresponding to goingout onto the first plateau. The width ry
for E = Bn is 30-40 MeV for all fissioning elements. Even if it is assumed that this value does not decrease for an
increase of 1 MeV in the excitation energy, for Pu240, with E = 4.9 MeV, we should have fr. f 5 MeV. When using
the expression for the level density based on the Fermi gas model:
where
Q (E, J) 2J+1 (E) exp
2 1/2:-E a3
.72_
9a2
(E) = 1/
12 (414 (K)514 exP [2 (aET/2]
(7)
(7')
[here E' = E ? A (the parameters a, a, and A were found by A. V. Malyshev from a comparison with the experi-
mental data for E = Bn)], calculation from Eq. (1) gives a hundred times greater values. Accordingly, we shall try
to analyze the experimental data, starting with the assumption B, and using Eqs. (1)-(7).
According to assumption B, the fissionablity reaches half its value at the plateau for T < Ef. We shall find
how the experimental values of the quantities T and SE are related to Ef and Ecurv in this case. Making the na-
tural assumption that in subbarier fission a role is being played by only one lower channel (for fixed values of J
and 11), and that the distribution of fission widths is described accordingly by x2, the distribution with one degree
of freedom [5], we find that in our case F' is equal to the mean partial fissionability:
, 1
F = S (a)
1+a '
and
St (a) = (1 + a) ( e_,'2 dx
1244
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
(8)
(9)
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
is the "fluctuation function" for the one-channel distribution,
as found in [6]. Differentiating (8) with respect to the energy,
using (1), (2), and (9), and assuming that the level density and
the radiation width change slowly with increase in energy in
/0 comparison with the rapid increase in the fission width, we
obtain:
Uri
1,0
10-2
40-3
10-4'
it,5 5,0
55 60 6',5 70
7,5
10-2
Ira
gn
4
5,0 55 6,0 5,5 70 7,5 E
Fig. 3. Energy dependence of the photofission
cross-sections of U238 (a) and Th232 (b).
found in the preliminary calculations which
widths.
or' _
OE
[S
Ecurv (a)-11.
(10)
For F' = 1/2, the solution of the transcendental Eq. (8) in
a gives the values a* = 0.37 and S(a*) = 0.675. Substituting
the experimental values IT ? in the left-hand
aE = SE
side of (10) and a = a* in the right-hand side gives the follow-
ing relation between SE and Ecurv:
EcuTV =- 56E ? 0,325 6E.
This shows that according to our analysis, the values of Ecurv
are a factor of 1.5 less than those found by the authors of [4]
when using assumption A. It should be noted that when making
calculations from formula (11) it is possible to use the values
of SE given in [4], since the fissionablity curves in assumptions
A and B, but with different values of Ecurv, are very nearly the
same, and the finite energy resolution changes them in the same
way.
To calculate Ef, , we use the condition:
Applying Eqs. (1) and (7) we obtain:
Ecurv r ao 1 .
E1 23t L 211Q (T) -
(12)
(13)
The results of the calculations made by Eqs. (11) and
(13) are given in the table. The values found for the "threshold
shift" Ef ?T are approximately a factor of 1.5 less than those
were made in [7] neglecting the fluctuations in the fission
Since, in the (d, pf) reaction, for a deuteron energy of 14 MeV, the neutrons captured by the target nucleus
can have a considerable orbital moment, the compound nucleus is formed in states with all possible combinations
of spin and parity. According to our assumptions, the plateau on the fissionability curve first shows up for ff?f ?
Accordingly, the subsequent rise in fissionablity cannot be accounted for by further increase in the fission width,
i. e., by opening up of new fission channels for compound nuclei, the fission probability of which was already very
large. The second rise accordingly means that at an excitation energy corresponding to the beginning of the first
plateau, for a considerable, and possibly even greater part of the compound nuclei, fission is still strongly pro-
hibited [there are at least two groups of combinations of the spin and parity of the compound nucleus for which the
thresholds are equal respectively to 41)and E(f2) (see table)], i, e., they differ by approximately 0.6-0.8 MeV. This
is also shown by the fact the value ofithe fissionability F at the first plateau is considerably less than unity. How-
ever, it is impossible to make a direct identification of the value of the fissionability at the first plateau with the
1245
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
fraction of the compound nuclei that fission through the lower barrier, since the height of the plateau may depend
on the spin of the target nucleus, as well as because of the possibility of an effect being exerted on this quantity by
anisotropy in the angular distributions of the protons and fragments. This last fact is also pointed out by the authors
of [4].
If it is assumed that Ai (see above) is several hundred kiloelectronvolts, the second rise in the fissionability
curve corresponds to fission of compound nuclei with negative parity. The consequences Of this assumption may be
compared with the experimental data on photofission, and fission by resonance neutrons.
Photo fission
The energy dependence of the photofission thresholds and the angular distributions of the fragments have been
investigated by many authors both with monochromatic 7-rays [8-10], and with a bremstrahlung spectrum [11, 12].
Those that have been best studied are the nuclei Th232 and U238. In heavy nuclei, y-quanta with an energy of 5-7
MeV experience only electric dipole and quandrupole absorption, which leads, for even-even target nuclei, to the
formation of compound nuclei in the states 1- and 2+. The photofission cross section is equal to
2+
? (2i-)/ r f \ (1-) / \
V'f?Cry rr-prc/,+ \r1?+re/ (14)
where a(2+) and a (2-) are the cross-sections for quandrupole and dipole photoabsorption respectively, and rc is the
total width of the Lcay processes, competing with fission. Below the (y, n) reaction threshold, we have r = Tc.
From general considerations, the dipole absorption cross section will be considerably greater than the quadrupole
cross section. If this is true, then, for a large difference between the thresholds in favor of quadrupole photofission,
the curve giving the energy dependence of the photofission cross section, with the ordinates plotted to a logarithmic
scale, should be qualitatively a straight line with the slope 27r/E(cilv for E < 41.), and a straight line with the slope
21t/E(2)rv for 41) < Ey < 42). The change in slope occurs at the point where ?he quadrupole photofission cross sec-
tion reaches situration, for f.(c2+)=Y Since, qualitatively, the curvature at the peak and hence the corre-
sponding value of Ecurv will be greater for a higher barrier, the above change in slope will be in the direction of a
decrease. Experimental curves for Th232 and U238, taken from [12], are shown in Fig. 3. It may be seen that the
form of the curves is qualitatively in complete agreement with our assumptions, especially for Th232, in which the
two linear segments are clearly visible. The values of E(clu)rv_for both nuclei may be found directly from the slope,
and are equal to 0.41 for Th232, and 0.68 for U238. If it is assumed that fr at the break point, Eq. (13) may
be used to find the values of the quadrupole photofission thresholds. They are equal to 5.7 and 6.0 MeV respectively.
For Th232, it may also be found that E(c2ilry = 0.68 MeV. To find the position of the second threshold, it is necessary
to know the total cross section for dipole absorption of y-quanta in this energy range. A lower limit for this quan-
tity may be found if it is assumed that the photofission cross sections reach saturation at Ey = Bn. Then:
ay
(1-) (2+) + ay = ayf (Ba).
(15)
It is known from the experiments of [8] on the angular distribution of the photofission fragments that at
EY = 6.14 MeV for U238, fission through the 2+ channel makes up about 20% of the total photofission cross section.
Judging from the curves in Fig. 3, the contribution from quadrupole fission in Th232 is even less. Then, using Eq. (15),
and assuming that the photoabsorption cross section changes only slightly as Ey changes by several hundred kilo-
electronvolts, it is possible to construct a curve for the fissionability through the 1- channel in the range E = 5.5-6
MeV, and determine 42) by means of Eq. (13), The results are given in the table.
It should be noted that the contribution made by the quadrupole component to the angular distribution of the
photofission fragments was first observed in [11],'while making measurements with a bremsstrahlung spectrum with
the limiting energy E = 9.3 MeV, and was found to be equal to 10%. If our values for the partial photofission
cross sections are averaged over a bremsstrahlung spectrum of this type, we get precisely the ratio cr-ra-li ,,10%
Y Y
For U234 and U236 nuclei, the photofission cross section has been measured in [10] with monochromatic y-radiation
at only the two points: Ey = 6.14 MeV and Ey = 7.0 MeV. For U234, 0,1 (6, 14) = 5+p mbarn and a j(7.0)= 52? 16
mbarn. The second point already lies higher than the binding energy ot the neutron, and no definite conclusion
can be drawn as to the position of the threshold of the channel
1246
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Our assumptions as to the mutual location of the 1- and 2+ thresholds lead to certain conclusions for which there
are not sufficient experimental data to check. Thus, for example, for E7 < E(11), as we get further down into the
subbarrier region, the angular distribution of the fission fragments shows an ever increasing role being played by the
quadrupole component, observation of which requires measuring the fragment yields at an angle of 450 to the beam
of y -quanta as well as at the angles 0 and 900. This has been pointed out in [18].
Fission by Resonance Neutrons
Experimental data are known which may be used to make a direct check of the Bohr-Wheeler formula. These
are the results of experiments on the resonance structure of the neutron cross sections. Such data exist for the four
fissionable nuclei U233, u235, pu239,, and Pu241.
Several of the first resonances in each of the above isotopes are divided by multilevel analysis into two groups
with different spins, and, for each group, the mean values are found of the distances between levels TY/, and of the
fission widths. If the Bohr-Wheeler formula (1) is correct, the "effective number of channels:"
H
Vje'fil? =- P (Ei; E)?
i=1
(16)
may be expressed in terms of the quantities observed experimentally. We shall now verify whether or not the values
of veff obtained in this way for different nuclei agree with our assumptions as to the structure of the fission channels.
1, U233 + n. Ground state 5+/2. Accordingly, the resonances correspond to the levels 2+ and 3+ of U234. If
the nucleus is axially symmetric at the saddle point, the rotation band of positive parity, which, according to our
assumptions corresponds to the lower fission channel, contains only the level 2+, while the level 3+ is a single par-
ticle level. The lower fission threshold of U234 (see table) lies more than 1 MeV below the neutron binding energy.
At E = Bn, the one particle channels will already be making a contribution, and, hence, v2e+ff > 74+ff 1. The ex-
perimental values calculated from our papers [13, 14] are v2e+ff ps 2 and ve3+ff 0.7.
2. U235 + n. The ground state of U235 is 7-/2, and the compound nucleus fissions from the states 3- and 4-.
In U236, only the position of the first threshold is known, which lies approximately 0.4 MeV beldw the neutron bind-
ing energy (see table). The rotation band of negative parity, which contains the level 3-, will lie 0.6-0.8 MeV
above the lower threshold, i. e., in the vicinity of E = Bn,, Also near Bn, several hundred kiloelectronvolts higher,
will be the first channels corresponding to single-particle excitations. Accordingly, the relation v4e-ff < ve3-ff < 1
will be approximately satisfied. According to the results of [15], the values of veff are equal to 0,6 and 0.15 for
the two systems of resonances. The values of the spins in these groups have not been determined experimentally.
It may be seen that our assumptions will give better agreement with the experimental data if it is assumed that the
levels with the larger fission width have spin 3.
3. Pu241 + n. Ground state of Pu241, 5+12. The fission widths have only been measured for nine resonances
[16], and the position of the fission thresholds is unknown. The values of veff are equal approximately to 1.5 and
0.2. If all our assumptions are correct, this means that the broad resonances correspond to J = 2, while the narrow
resonances correspond to J = 3, and the threshold of the lower channel is located approximately 0.5 MeV below the
neutron binding energy. A study of the (d, pf) reaction in the Pu241 nucleus is of particular interest in this respect.
4. PU239 + n. Ground state 1+12. The mean distance between resonances is -17) = 2.9 eV, which corresponds
to ID9+ = 11.6 eV, Di+ = 3.9 eV. According to the most recent measurements [15], II+ 40 MeV, fr 160 MeV,
and,hence, ve9+ff = 0.26, v9e+ff = 0.02. Thus, the assumptions as to the structure of the fission channels, which, in
other cases, are in complete agreement with experimental data, here lead to an error of two orders of magnitude.
This inconsistency may be removed if it is assumed that the ground state of the Pu239 nucleus has negative parity.
There is no 0- level in rotation bands of negative parity, and the channel corresponding to it is necessarily a single-
particle one. If it lies approximately 2.0 MeV above the lower threshold, this will give the small penetrability ob-
served. The fluctuations in the fission widths of the PU239 resonances are well described by the one-channel Porter-
Thomas distribution [5]. This also shows that the threshold for fission of PU239 by S-neutrons is located high.
In [17], a systematic scheme was proposed for the parities of the ground states of a-active nuclei. The posi-
tive parity of PU239 is one of the very few places where the scheme differs from the experimental data.
1247
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
CONCLUSIONS
It has been assumed in this paper that the Bohr-Wheeler formula (1) gives an order of magnitude description
of the mean fission width. In analyzing the data on the (d, p5) reaction, it follows unambiguously from this assump-
tion that, first, there are at least two sets of combinations of spin and parity of a fissionable nucleus for which the
fission thresholds differ by 0.6-0,8 MeV, and, second, these thresholds are located higher than had previously been
assumed. The data on the (y, 5) reaction were analyzed under the additional assumption that the photoabsorption
cross section is only slightly dependent on the energy in intervals of the order of 1 MeV, as compared with the ex-
ponential increase in the fission width in the range Ey = 5-7 MeV. Our treatment also leads to appreciably higher
values of the photofission thresholds than those accepted up to the present time, with the fission barrier in quadrupole
photoabsorption located 0.6-1.0 MeV below the dipole photofission barrier. In comparing the results of analyzing
the (d, pf) and (y, f) reactions, it is natural to assume that the first rise in fissionability in the (d, pf) reaction cor-
responds to channels with positive parity, while the second corresponds to channels with negative parity. All these
conclusions are in agreement with A. Bohr's hypothesis as to the structure of the fission channels of even-even nuclei,
if the distance between the rotation bands of positive and negative parity is Ai Ftc 0.6-1,0 MeV.
In accordance with the results of treating the (d, p5) and (y, 5) reactions, and within the framework of
A. Bohr's hypothesis, a distribution of the fission channels may be given for which the Bohr-Wheeler formula gives
a quantitative description of the experimental data on the mean fission widths of the resonances of the (n, f) reac-
tion, except for the data on the PU239 nucleus. To get rid of the large difference noted in this last case, it must be
assumed that the ground state of the Pu239 nucleus has negative parity.
LITERATURE CITED
1. D. Hill and J. Wheeler, Phys. Rev., 89, 1102 (1953),
2. N. Bohr and J. Wheeler, Phys. Rev., 56, 426 (1939),
3. A. Bohr, In the book: "Materials from the International Conference on the Peaceful Uses of Atomic Energy
(Geneva, 1955)," Vol. 2 [Russian translation], Moscow, Fizmatgiz (1958), p. 176.
4. J. Northrop, R. Stokes, and K. Bayer, Phys. Rev., 115, 1227 (1959).
5. C. Porter and R. Thomas, Phys. Rev., 104, 483 (1956).
6, A. Lane and J. Lynn, Proc. Phys. Soc., 70, 557 (1957).
7. L. N. Usachev, V. A. Pavlinchuk, and N. S. Rabotnov, ZhgTF, 44, 1950 (1963).
8. B. Forkman and S. Tohansson, Nucl. Phys., 20, 136 (1960).
9. H. De Carwalho, A. Manfredini, and M. Muchnik, Nuovo Cimento, 25, 136 (1960).
10. J. Huizenqa et al., Nucl. Phys., 34, 439 (1962).
11. A. I. Baz' et al., In the book: "Transactions of the Second International Conference on the Peaceful Uses
of Atomic Energy," Papers presented by Soviet scientists, Vol. 1 [in Russian], Moscow, Atomizdat (1959),
p. 362.
12. L. Catz, A. Baerg, and F. Brown, 13/200, Proc. of II?d UN Conference on PUAE (Geneva, 1958), Vol. 15
(1958), p. 183.
13. C. Reich and M. Moore, Phys. Rev., 118, 718 (1960).
14, E. Vogt, Phys. Rev., 118, 724 (1960).
15. I. V. Kirpichnikov et al., "Atomnaya energiya," 16, 110 (1964).
16. A. Simpson and M. Moore, Phys. Rev., 123, 559 (1961).
17. V. N. Andreev, ZhETF, 42, 913 (1962).
18. J. Griffin, Phys. Rev., 116, 107 (1959).
1248
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
NEUTRON ANGULAR AND ENERGY DISTRIBUTION
AT THE BOUNDARY OF TWO MEDIA
(UDC 539.125.52)
V. A. Dulin, V. G. Dvukhsherstnov, Yu. A. Kazanskii,
and I. V. Shugar
Translated from Atomnaya tnergiya, Vol. 17, No. 6,
pp. 486-492, December, 1964
Original article submitted August 8, 1963; in revised form, July 2, 1964
Measurements were made of the angular and energy distributions for fast neutrons (0.4-3.4 MeV)
emerging from water, graphite, aluminum, iron, nickel, and lead at a water-medium interface
after penetration of thicknesses equalling 1.5-4.6 mean free paths. The reaction D(D, n)He3 was
the neutron source. Measurements were made with a single-crystal fast neutron scintillation spec-
trometer having y-ray discrimination. The results of the measurements are compared with cal-
culations.
The most complete characterization of a neutron flux in a medium is the angular and energy distribution.
An exact solution of the transport equation for the purpose of defining this characterization is fraught with great
difficulty. Approximate methods require experimental verification, and all the more so if the system of constants
is not complete. Combined experiments are of interest since they furnish recommendations which are important
for practical shielding construction; in addition, they can serve
as criteria for the correctness of a selected approximate method,
and, finally, they enable one to refine the system of constants.
The intensities of existing monoenergetic sources and the
efficiencies of fast neutron spectrometers were unsatisfactory,
until recently, for detailed studies of the neutron flux in materials.
High-efficiency neutron detectors [1, 2] have made it possible to
carry out measurements of angular and energy distributions for
the case of a point isotropic, 3.2 MeV source at distances of
1.5-4.6 mean free paths and in the angular range 20-70?. This
work is a continuation of work started in [3].
Deutron 200 I
beam
\ Experimental
med i urn
Fig. 1. Diagram of experimental apparatus.
Experimental Arrangement
A cascade neutron generator was the neutron source. Deu-
terons with an energy of 400 keV were incident on a Zr-D target
200 keV thick, and thus there were neutrons with energies of 3.08-
3.4 MeV along the deuteron beam. A diagram of the experimental
apparatus is shown in Fig._1. Selection of a neutron beam at a
definite angle was accomplished by using a conical collimator
with an angular resolutions of ? 8?. In order that neutrons which
emerged from the experimental material in the vicinity of point
A in the 0? direction not be scattered in water, there was an air
space at the boundary. The neutron background was recorded
with a collimator filled with water. About 1% boric acid was
added to the water, and the spectrometer was enclosed on the end
and sides with a 1.5 cm thick layer of lead. In this way, the ratio
1249
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
0(E0)
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
between the flux of 7-rays and neutrons with the collimator
open did not exceed 10 even at an angle of 70?. In allthe
experimental media, the ratio of neutron signal to back-
ground was about 5-6 and 20-30 for angles of 70? and 20?.
200 respectively.
In measuring the energy distributions outside the ex-
perimental media, the detector was located in the air space
close to the point A. Neutron spectra were measured with
700 the fast neutron scintillation spectrometer previously de-
scribed [4]. An photomultiplier was used in the
spectrometer, Discrimination against 7-rays was ac-
complished by a circuit already described [1], the only
differences being that D-10 diodes were used in the dynode
and Anode circuits, and after combining pulses, the nega-
Fig. 2. Neutron energy distribution at 0? with no tive cotnponent of the resulting pulse was shorted to ground
experimental Medium, through a D-11 diode. The spectrometer threshold equalled
0.5 MeV for a 103 attenuation of the 7-ray count. The
pulse height distribution Was Measured with a one hundred channel analyzer. The energy scale of the spectrometer
was set; and periodically checked, by measurements of the pulse height distribution at the 0? angle. Instrumental
stability (chiefly that of the phototnultiplier) was 286/oduring the contse of all the Measurements. The pulse height
dittribution was measured three to five times at each angle.
Conversion of the pulse height distributions to energy spectra was carried out by means of the formula
2 3 Fn, MeV
ct? (En) = ?
En dV d.[ dV I
1 ?exp [?I (En) di dE X dV LN iv dE J '
(1)
where t (En) is the macroscopic neutron scattering cross section of hydrogen in the crystal; d is the crystal thick-
ness; dV/dE is the derivative of pulse height with respect to energy.
Thus the analysis of results was reduced to a multiplication of the measured pulse height distribution N(V)
by dV/dE (each channel), differentiation with respect to the pulse height scale by the "sliding zone" method, fol-
lowed by multiplication by dV/dE and correction for detection efficiency. It has been shown that differentiation
by the sliding zone method makes it possible to use reasonably Small steps(... 70 keV), and this does not lead to
"oscillation" of the results.
As is Well known, differentiation of the instrumental spectrum is not a rigorous method of analysis since the
shape of the line from the crystal differs from a "plateau" (dE/En). However, as V. G. Zolotukhin has shown, this
is unimportant for continuous spectra even when working with crystals of large diameters. Results which were ob-
tained by the method deScribed for the neutron spectrum at 0? with no experimental medium are shown in Fig. 2.
It is apparent that there is a weak maximum in the spectrum at ? 2.5 MeV which is principally the result of not
considering the actual shape of the line from the crystal.
Discussion of Results
Fast Neutron Angular and Energy Distributions. Fig. 3 gives the experimentally obtained
neutron energy distributions at angles of 20, 40, and 70?, and Fig. 4 gives the angular distributions for neutrons in
the energy groups 0.4-1.2, 1.2-2.0, 2.0.2.8, and 2.8-3.4 MeV after penetrating layers of water (3.3 Xt), graphite
(3.3 Xt), aluminum (1.5 Xi), iron (3.0 and 4.6 Xi), nickel (2 and 4 Xi), and lead (3 Xt). All thicknesses are given
in mean free paths for 3,4 MeV neutronS. It is clear from the figure that the shape of the energy spectrum varies
irregularly in all media. But in every case except that of water, the elastic scattering peak appears rather clearly
(in iron, a peak caused by inelastic scattering appears), and softening of the spectrum with increasing angle is ob-
served. To a first approximation, the dependence of the energy groups on angle is the same for all media. The
length of path also has little effect on the angular dependence of the energy groups.
In Fig. 5, the angular distributions for the 2.8-3.4 MeV group are shown along with results from a single-
scattering approximate calculation which was done with isotropic, and actual anisotropic, scattering cross sections.
1250
Declassified and Approved For Release 2013/03/14 : CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
P(E0)
2
0
2
1
0
2
1
1120
1
'LLAI,
Pe (4,6 At)
Ni (4 At)
1
..rm-
_
...
1 .ii.i?
23
Al
Pb
1-
?
-
i
i
1
1
0
2
3
0
1
2
3 En, MeV
Fig. 3. Neutron energy distributions in water, graphite, aluminum, iron, nickel, and lead
at angles of; 1) 200; 2) 40?; 3) 70?.
The experimental and calculated angular distributions were multiplied by sin O. The diagreement between experi-
ment and theory for the 2.8-3.4 MeV group indicates the importance of the role of multiple scattering.
For practical purposes, the angular distribution of neutron dose can be extremely useful. The angular distri-
bution of dose for neutrons with energies of 0.4-3.4 MeV multiplied by sin 0, i. e., sin 0 (E , 0) D (E) dE
0:4
is shown in Fig. 6.
The dependence of dose on energy D(E) was taken from [6]. Data for the angles 0-20? can be obtained from
single-scattering approximate calculations for the 2.8-3.4 MeV group and by means of extrapolation for the re-
maining energy groups. As can be seen from Fig. 6, the magnitude of the dose in the range 20-70? for all media
except water depends slightly on angle, and can be described by the formula
r 0
D (13) 0 "P
where 27r/3 00 27r. For water, the dose dependence on angle is close to that for 7-rays [7] (e,
(2)
1251
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
17-Z000Z1-00900n1961-Z0-01-c1C11-V10 :171-/?0/?1,0z eseeiej Jod panaiddv pue Pe!PsseloeCI
10
10 -
Fig. 4.
F (.4,6?
20 30 40' SO 60 70
C.
(2,1t
pi: t1
I:I. r I. t U I. 1'
10 30' 40 50' 60 70 0' 21J 30' 40 50' 60 70 a 20 30 40: 50, 60' 70
6,.. degrees;
Neutron angular distributions in water, graphite,; aluminum, iron, nickel, and lead per unit solid angle for the energy groups:.
MeV.;. .); 2,0-2,8 MeV; 11) 1..2.-2:?0 MeV; 0,), 0,4-1..2 MeV.
0) 2.8-3.4
17-Z000Z1-0090001961-Z0-01-c1C1I-V10 icoio eseeiei -101 panaiddv pue Pe!PsseloeCI
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
100
1,0
gl
1,0
C(3,3.1,)
?
_
'MI
I
I
I
11(11
FepAt)
1
.
?
Pb(3)
1 1 I ,
1
.
I
.
i
?
I
i
'lilt
Fe(4,6A,)
? .
?
.
0 10 20 30 40 50 60 70 0 10 20 30 40 50
0, degrees
Fig. 5. Angular distribution of the 2.8-3.4 MeV group for graphite,
load, and iron in the solid angle dsl: 1) single-scatter approximate
calculation assuming isotropic differential scattering cross section;
2) single-scattering approximate calculation using actual differential
scattering cross sections for neutrons with energies close to 3.4 MeV.
Pb 3c)
Ni 44 , )
AN (2 4
Pe (4 6
Fe (3 4
11111.11110
Al
MOM
MI
MI P11111111
H20
o 10 20 30 40 50 60 70
0, degrees
Fig. 6. Angular distribution of dose
in the solid angle dl) for 0.4-3.4 ?
MeV neutrons.
Fast Neutron Energy Distributions
The fast neutron energy distributions at the boundary of two media
43(E0, E, R) are not only of interest in themselves, but also may be used
for checking the angular energy distributions obtained, In Fig. 7, there
are shown neutron energy distributions both measured at the water-
medium boundary (in the space A) and obtained by integration of angular
energy distributions. Extension of the results to the region of angles
greater than 70? was obtained by exponential extrapolation. Extrapola-
tion into the range of angles 0-20? for the 2.8-3,4 MeV group was made
in accordance with the results from a single-scatter approximate calcu-
lation. The energy distribution of neutrons in the space A is the sum of
scattered and unscattered radiation. The integral over angle only takes
into account scattered radiation. Measurement of the energy distribution
at 0? makes it possible to determine the contribution of unscattered radia-
tion. The satisfactory agreement between measured spectra and the
spectra obtained by integration indicates that the choice of exponential
extrapolation was a good one and apparently one that reflects the main
features of the angular energy distributions, Results are also given in
Fig. 7 for moments method calculations of the energy distributions in
water, graphite, and iron with an isotropic point source of neutrons [8]
and for Monte Carlo calculations [9] in iron with an effectively plane,
monodirectional source of 3.0 MeV neutrons. The same figure gives the
1253
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
1254
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
4-1
Fig. 7. Neutron energy distributions: ???) measured at the water-medium boundary (in space A);
---????) angular energy distributions obtained by integration with allowance for unscattered neu-
trons; ?) results of calculations in water and graphite [8]; -----) (histogram) calculation [9];
0) results of measurements in shield geometry [10].
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
results of fast neutron measurements with a single-crystal scintillation spectrometer under shielding geometry con-
ditions and for an effectively plane, monodirectional source of ? 3 MeV neutrons [10]. A detailed comparison of
the energy distributions is difficult because of the differences of initial energies and of geometries. However, the
differences between the energy distributions is not very great even with such differing geometries.
The authors consider it their pleasure to express thanks to S. G. Tsypin for valuable remarks and discussions
and to L. A. Trykov for helpful advice. In addition, the authors-thank A. P. Klimov for maintaining successful opera-
tion of the neutron generator and A. T. Bakov for help with the measurements.
LITERATURE CITED
1. F. Brooks, Nucl. Instrum. and Methods, No. 4, 151 (1959).
2. R. Owen, Atomnaya tekhnika za rubezhom, No. 1, 16 (1960).
3. V. A. Dulin, Yu. A. Kazanskii, and I. V. Shugar, Atomnaya energiya, 14, 488 (1963).
4. V. A. Dulin, et al., Pribory i tekhnika eksperimenta, No. 2, 35 (1961).
5. K. Laniosh, Practical Methods of Applied Analysis [in Russian], Moscow, Fizmatgiz (1961), p. 327.
6. Kh. D. Androsenko and G. N. Smirenkii, Pribory i tekhnika 6ksperimenta, No. 5, 64 (1962).
7. Yu. A. Kazanskii, Atomnaya energiya, 8, 432 (1960).
8. H. Goldstein, Fundamentals of Reactor Shielding [Russian translation], Moscow, Gosatomizdat (1961).
9. M. Leimdorfer, FOA 4A 4196-411 (May, 1961); AB.Atomenergi, Internat. Repots RFA--88 (1962).
10. G. During, R. Jansson, and N. Starfelt, Symposium on the Detection, Dosimetry, and Standardization of Neu-
trons (Harwell, 10-14, December, 1962), Vienna, IAEA, SM-36/102.
1255
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
THE NEUTRON BACKGROUND AT THE SURFACE OF THE EARTH
(UDC 537.591: 550.35)
G. V. Gorshkov, V. A. Zyabkin, and 0. S. Tsvetkov
Translated from Atomnaya Energiya, Vol. 17, No. 6,
pp. 492-496, December, 1964
Original article submitted December 24, 1963
Measurements were made of the slow neutron flux above soil and water surfaces. The slow neutron
flux above soil was 3.1 times greater than that above a water surface. Possible reasons are discussed
for the increase in slow neutron flux above soil as compared with water. The thermal neutron flux
was computed at the surface of a material in which fast neutrons are formed by cosmic rays.
It was concluded that the difference in the slow neutron fluxes above soil and water is caused
to a considerable degree by the cosmic ray generation of neutrons in the upper layers of the earth's
crust.
The curve for the altitude dependence of the neutron flux produced by cosmic radiation is distorted close to
the surface of the earth by a number of factors, chief among which are the following:
1. Cosmic ray production of neutrons in the upper layers of the earth's crust.
2. Distortion of the neutron spectrum by the surface of the earth. In 1940, this factor was pointed out by
Bethe, Korff, and Placzek [1] who considered the case of an infinitely extended water surface. A part of the fast
neutrons produced in the atmosphere close to a water surface are incident on the water, and are moderated in it
to thermal energies. Some of these thermal neutrons, diffusing back into the air, increase the thermal neutron
density close to the water surface in comparison with that in the free atmosphere where the relative thermal neu-
tron density is considerably lower because of capture by nitrogen in the N14(n, p)C14 reaction (formation of radio-
active carbon).
Since the efficiency of any neutron detector depends on the neutron spectrum being recorded, a distortion
of this spectrum can influence the results of a measurement.
3. Production of neutrons in the earth's crust as a result of natural radioactivity.
In this paper, we attempt to estimate the slow neutron flux at sea level above water and soil surfaces.
Experiment
Neutron radiation was recorded with a scintillation counter which consisted of a slow neutron detector and
photomultiplier 200 mm in diameter.
The detector, prepared by us of enriched boric acid (enriched in B11) to 91/0) and zinc sulfide by the method
of joint fusion [2] with consideration of the changes mentioned in [3], was in the form of a 200 mm disc. The con-
struction of the scintillation counter is briefly described in [4, 5]. Detection efficiency was determined with a
thermal neutron flux which had been measured by activation of indium foils and of manganese (in the form of KMn04)
with subsequent radiochemical concentration of the radioactive manganese by the Szilard-Chalmers method.
The absolute number of In116 and Mn56 decays was determined through the y-radiation of these isotopes by
means of a scintillation y-counter calibrated by the 6-y coincidence method [6].
The slow neutron flux at sea level was measured at the following points:
1. in the environs of Leningrad at Zelenogorsk, approximately 3 km from the Gulf of Finland;
2. on a pier in Zelenogorsk (soil-water boundary);
1256
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Results of Neutron Background Measurements at the Earth's Surface
Point of
observation
Type of
measurement
Number of
pulses recorded
Counting time,
h
Counting rate,
cts/h
Signal difference with
and without cadmium
cts/h
neutsicrn2. day_
1
2
3126
Without cadmium
With cadmium
Without cadmium
With cadmium
Without cadmium
With cadmium
5287
1435
1204
718
2276
1246
8.92
7.13
3.09
3.92
7.16
6.5
593 ? 8
201 ? 5
391 ? 11
205 ? 7
318 i 7
192 ? 5
392 ? 10
186 ? 13
? 9
285 ? 7
136 ? 9
93 ? 6
3. above the surface of the water in the Gulf of Finland approximately 1 ktri from shore. Measurements were
made from June 29 to July 2, 1960.
The absolute height above sea level did not exceed 30 m. Atmospheric pressure during the measurements
underwent only insignificant changes (1013-1018 mbar).
For measurement of the neutron flux above the water surface, the apparatus was installed in a light, wooden
boat. The depth of the gulf at the point of measurement was about 5 m. For all measurement points, the neutron
detector was placed 1 m from the surface.
All neutron radiation measurements were carried out by the difference method using cadmium shields which
completely surrounded the counter. From the difference in counter readings with and without the cadrriium shields
(1 mm thick), the net effect produced by slow neutrons was determined. The result obtained With the Counter co-
vered by cadmium was taken as background. By a special experiment, performed in a deep uridergrotind location
it was demonstrated that the detector background produced by natural contamination with a-active element was
(6.1 ? 0.1) cts/cm2. day. Results of the measurements are given in the table,
The value obtained for the slow neutron flux at sea level in the vicinity of the'ground [(285 ? 7) neuts/cm2. day]
is in excellent agreement with previous results [7-9] (240, 230, and 290 neuts/cm2? day, respectively). The value of
the slow neutron flux above a water surface [(93 ? 6) neuts/cm2 day] is in agreement with other results [10][(78 2)
neuts/cm2? day].
The main feature of the results is the fact that the slow neutron flux above the surface of the ground (see table,
point 1) is 3.1 times greater than the neutron flux above a water surface (point 3). The magnitude of the flux at the
water-ground boundary has an intermediate value.
It is impossible to explain the increase in the value of the slow neutron flux close to the surface of the ground
by a contribution from neutrons of purely terrestrial origin, i. e., because of (a, n) reactions, spontaneous fission of
heavy nuclei, etc. As has been shown [4], the slow neutron flux measured in underground workings (in granite and
marble, for example) at a depth of about 200 rn water equivalent is (13 ? 2) and (6 ? 1) neuts/cm2. day; respectively,
which is 5 and 2% of the magnitude of the slow neutron flux at the surface of the ground.
Measurements at other points on the earth's surface have shown that the slow neutron flux which is obtained
at the same geographic latitude (60? N) at sea level and normalized to a pressure of 760 mm Hg exhibits some varia-
tion (approximately 200-300 neuts/cm" day) depending on the chemical composition of underlying rock, the mois-
ture content and the topography of a location, and the presence of surrounding objects.
In addition, the neutron flux at point 1 was also measured in winter with a snow cover present. With snow
cover present, it was shown that the slow neutron flux, normalized to a pressure of 760 mm Hg, was 1.65 times less
than the neutron flux measured at the very same spot during the dry summer seaSon.
The reduction in slow neutron flux with the transition from a soil to a water surface was observed in [11] and
also in [12, 13]. According to the data in [13], the slow neutron flux in firn areas of the high mountain regions is
1257
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
I 1 ?
Diagram for calculating the rate of
thermal neutron production.
Theoretical Considerations
reduced by a factor of 1.4 in comparison with the flux at granite out-
croppings in the same area. The neutron flux at granite intrusions in
the high mountain regions is several times greater than the neutron
flux at a glacier located in the same area [12].
It is impossible to explain the reduction of the slow neutron flux
above water as compared with that above rock by a distortion of the
energy spectrum of the neutron component of the cosmic radiation
since, in that case, the slow neutron density above water surface should
be greater than the neutron density above rock. As indicated above,
the slow neutron intensity of purely terrestrial origin in rock is also in-
sufficient to explain this effect.
The next factor that one must take into account is neutron pro-
duction through the action of cosmic radiation in underlying rock and
in water. We shall consider whether the strength of this additional
source of neutron formation at sea level is sufficient to exert a notice-
able effect on the magnitude of the neutron component of cosmic
radiation.
We compute the thermal neutron flux per unit area at the surface of an infinitely extended material in which
cosmic rays are producing fast neutrons. To simplify the calculations, we assume: 1) the cosmic radiation is nor-
mally incident on the surface of the material; 2) there is a vacuum above the surface of the material; 3) there is
no fast neutron capture in the material.
We introduce the following notation: I is the thermal neutron flux per unit area at the
cm2-sect; qc, is the rate of fast neutron production by cosmic rays in the given material at
p is the density of the material, gicm3; xc is the absorption length of the neutron-producing
surface of the material,
its surface, g'-sec';
cosmic ray component
in the given material, cm; L is the fast neutron slowing down length in the given material, cm; 1 is the thermal
neutron diffusion length in the given material, cm; D is the thermal neutron diffusion coefficient in the given mate-
rial, cm.
We compute the rate of thermal neutron formation per unit volume of the material at the point 0 at a depth
h(see figure) because of slowing down of fast neutrons produced by cosmic rays at the point P in a volume element
dV of the material.
The rate of fast neutron production q in a volume element dV = r2dr sin e de thp located at a distance PK =
h?r case, from the surface of the material will be a factor exp[?(h?r cose)/xc] less than at the surface. This is
explained by the absorption of the neutron-generating component in the layer of material, i. e.,
h?r cos 0
q = qo dve x
The fast neutron flux through unit area at point 0 produced by fast neutrons arising from the volume element
dV at a distance PO = r from the point 0 will be
The number of fast neutrons emitted at point P in the volume element dV and converted to slow neutrons per
unit volume at point 0 will be
r
4:r2 .
1258
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
(1)
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
This expression gives the rate of formation of slow neutrons per unit volume at the point 0 because of the
slowing down of fast neutrons emitted from the volume element dV at Pt
h?r cos 0 r
Q (h, r, 0, cp) = 4nr2p-1E qc0X e-71- 1 r2 dr sin 0 (10 dcp.
The rate rate of formation of slow neutrons per unit volume at point 0 because of fast neutron S Which are produced
throughout the entire thickness of material will be
h n/2 h/cos 0 ' (3 1
2nco r (cos 0 i 1
Q (h) = 41n-Q-9.1.-- , dy fe ''''C [ sin 0 de c er (DOSxc ? L) di. + s
sine de e xc L )
A/2 h
sin 0
e it jt
sin 0 d0 1
_ 1? 7c [
, L cos 0? k c ?
L cos 0 do e xe
L cos 0 ?xc _I ?
0 0
1}
In an unbounded medium, the thermal neutron flux at a distance h from an infinitely thin layer of thickness
dh in unit volume of which Q thermal neutrons are formed per unit time is according to diffusion theory,
Q1 -
ainf (h) ?
2D et ("a'
For approximate calculations, one can consider that the thermal neutron flux at the material-vacuum boundary
is reduced by a factor of two in comparison with the flux in an unbounded medium at the same distance from an in-
finitely thin layer since the thermal neutron flux from the vacuum into the material is t etc).
Then we have for the thermal neutron flux at the surface of the material from ad infinitely thin layer of thick-
ness dh located at a depth h
h
dl (h)=- eT dh.
4D
(2)
In order to determine the total thermal neutron flux at the surface of the material, it is necesSary to integrate
expression (2) over the entire thickness of the material, considering further that the rate of slow neutron formation
per unit volume also depends on depth:
oz, h
I=, Q? (hi` e- dh
4D
n/2
_
qovel 1 c' sin 0- dO C -
cc' h (L clos 0+ 11 ) dl
z_
8D 1 L cos 0? x I e
o c b
n ex, 1 i
c'
?h (---1--, sin 0 de e xc ' ) dh? } .
L cos 0?x
c b,
0
The final result for the thermal neutron flux at the surface of the material takes the form
I = goOc12
8D L (xc+ 1) 1 lri(1+
xc
(3)
In order to estimate the thermal neutron flux above water and granite resulting from the generation of fast
neutrons in these materials by cosmic rays, one needs to know the slowing down length for the fast neutrons which
are produced in water and granite by cosmic radiation. The exact energy spectrum of these neutrons is unknown.
In [14], the mean neutron energy for neutrons produced in lead and carbon by cosmic rays was estimated to be
approximately 3 MeV. We used the slowing down length for fast neutrons with an energy of 3 MeV.
1259
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Water Surface: go = 2.89-10-5g-1-sec-1[10]; p = 1 gicm3; xc = 169 g/cm3[10] or xc = 169 cm;
= 2.3 cm [15]; L = 6.4 cm [16]; itr = 0.49 cm [17] (it, is the transport mean free path); D /tr/3 = 0.49/3 =
0.16 cm.
Substituting these values into formula (3), we find I = 1.72. 10-5 neuts/cm2- sec 15 neuts/cm2. day.
Granite: p = 2.78 g/cm3[15]; 1 = 10.3 cm [15]; D = 1.03 cm [15]; L = 45 cm (our calculation).
The mean atomic weight of granite is 27.6 [15], i. e., close to the atomic weight of aluminum -27. For the
estimate, we assumed that the magnitudes of go and xc were equal to the corresponding values for aluminum:
go = 5.2. 10-5 g'-sect [18]; xc = 220 g/cm2 [19] or xc = 220/2.78 = 79 cm,
From formula (3), we find I = 1.94.10-3 neuts/cm2. sec :z1 170 neuts/cm2. day.
It should be noted that the rate of neutron formation by cosmic rays in granite is only 1.8 times as much as
in water, but the thermal neutron fluxes resulting from their formation differ by an order of magnitude. This is ex-
plained by the differing slowing down and diffusion characteristics of granite and water.
CONCLUSIONS
The calculations that were made have shown that the production of neutrons by cosmic rays in the upper layers
of the earth's crust can exert a strong influence on the slow neutron flux at sea level.
In order for us to compare calculation with experiment, it would have been necessary to compute the slow
neutron flux at surface of the ground which arises from neutrons produced in the atmosphere. For such an estimate,
it is necessary to know the intensity and spectrum of atmospheric neutrons and the possible distortions of them at the
surface of the earth. We did not have such data available, and therefore such an estimate was not made in this
paper.
However, it is clear that the contribution of the neutrons which are produced in granite by cosmic rays to the
total neutron flux above its surface is considerably greater than in the case of water. Therefore, one can suppose
that the difference in slow neutron fluxes above water and soil surfaces are caused to a considerable degree by the
cosmic ray production of neutrons in the upper layers of the earth's crust.
LITERATURE CITED
1, H. Bethe, S. Korff, and G. Placzek, Phys. Rev., 57, 573 (1940),
2. T. V. Timofeeva and S. P. Khormushko, Izv. AN SSSR, Ser. fiz., 22, 14 (1958).
3. B. S. Grebenskii et al., idem, 25, 500 (1961).
4. G. V. Gorshkov, V. A. Zyabkin, and 0. S. Tsvetkov, Atomnaya gnergiya, 13, 65 (1962).
5. G. V. Gorshkov and 0. S. Tsvetkov, In Chemistry of the Earth's Crust, Vol. 2, Moscow, Izd-vo, AN SSSR
(1963), pc 409.
6. G. V. Gorshkov, A. N. Silant'ev, and 0. S. Tsvetkov, Radiokhimiya, 4, 244 (1962).
7, A. P. Zhdanov and A. S. Serdakov, Dokl. AN SSSR, 31, 861 (1941).
8. N. Kaplan and H. Yagada, Rev. Scient. Instrum., 23, 155 (1952).
9. K. Mather, Austral. J. Phys., 9, 147 (1956).
10. E. Bagge and S. Skorka, Z. Phys., 152, 34 (1958).
11. G. V. Gorshkov, S. P. Khormushko, and 0. S. Tsvetkov, Dokl. AN SSSR, 131, 933 (1960).
12. V. V. Cherdyntsev and V. I. Meshkov, Bulletin of the Committee for the Determination of the Absolute Age
of Geological Formations, No. 1, Moscow, Izd-vo, AN SSSR (1955), p. 61.
13. L. I. Shmonin et al., Uch. zap. Kazakhsk. un-ta, 30, No. 5, 25 (1957).
14. V. Cocconi-Tongiorgi, Phys. Rev., 76, 517 (1949).
15. E. M. Filippov, Applied Nuclear Geophysics [in Russian], Moscow, Izd-vo, AN SSSR (1962).
16. N. A. Vlasov, Neutrons [in Russian], Moscow, Gostekhteorizdat (1955).
17. Nuclear Physics Handbook [English translation], L. A. Artsimovich, edited, Moscow, Fizmatgiz (1963).
18. A. Tobey and C. Montgomery, Phys. Rev., 81, 517 (1951).
19, G. Benardini, G. Cortini, and A. Manfreclini, Phys. Rev., 76, 1792 (1949).
1260
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
ADDITION OF HETERO-ORGANIC COMPOUNDS TO POLYSTYRENE
(UDC 621.387.464: 678.746)
E. E. Baroni, S. F. Kilin, T. N. Lebsadze, I. M. Rozman,
and V. M. Shoniya
Translated from Atomnaya nergiya, Vol. 17, No. 6,
pp. 497-499, December, 1964
Original article submitted November 30, 1963
To obtain polystyrene containing Pb, Bi, Hg, Sn, and Se, the solubilities of their phenyl compounds
in the monomer are determined. The optimum conditions for thermal polymerization of styrene
containing these compounds are described in detail. Data are given on the preparation of polymeri-
zation products containing phenyl compounds of Pb (12%), Bi (1901o), Hg (33%), As (1201o), Sn (1101o),
and Se (1001o, by weight). The polymerization products are transparent and stable in air. The mate-
rial is intended for the production of scintillation detectors and y-ray and neutron shielding. Data
are given on the luminescent yield of plastic scintillators containing Pb, Hg, and Sn.
Scintillation methods of detecting neutrons, x-rays and y-rays make use of inorganic crystal phosphors, organic
monocrystals and liquid solutions [1-5], and also of plastic scintillators (PS) [6-9]. The atomic composition of PS
prepared by polymerizing solutions of organic phosphors in certain monomers (styrene and vinyltoluene) is limited to
the light elements C, H, and 0. These substances have very small coefficients of photoelectric absorption, which
causes considerable difficulty in the use of PS for 7-ray spectroscopy?though large PS are very efficient for de-
tecting low-intensity radiation,
Methods have been described for increasing the photoelectric absorption of plastics by incorporating heavy
elements in the form of organic compounds [10-12]. For example, 9.4 wt, 010 lead and 10% bismuth can be added
to polystyrene-based PS with 1,1', 4,4'-tetraphenyl-butadiene-(1,3) as luminous additive [10]. However, these poly-
merization products are unstable in air owing to the insufficient stability of the compounds used (bismuth and lead
hexahydrobenzoates, lead naphthenate, triphenylmethyl lead). An attempt to use Pb, Hg, or Bi isobutyrates or tri-
methylacetates, together with polycyclohexylmethacrylate in polyvinyltoluene, was unsuccessful [11].
We have performed experiments on incorporation into polystyrene-based PS of Pb, Bi, Hg, As, Sri, and Se in
the form of more stable and easily prepared compounds. The present paper gives test results on procedures for
thermal polymerization of styrene containing the phenyl compounds of these elements. [13] describe another meth-
od which we have developed for incorporating hetero-organic compounds in PS.
The incorporation of various chemical elements in PS is not only of interest for use in the measurement of
ionizing radiation, but is also important for the preparation of comparatively cheap 7-ray and neutron shielding
materials [14-16].
Preparation of PS with Organometallic Additives
We used the following hetero-organic compounds: tetraphenyl lead [17], diphenyl mercury, tetraphenyl tin,
triphenyl arsenic, triphenyl bismuth [18] and diphenyl selenium [19]. The best method of purifying these compounds
(except diphenyl selenium) is sublimation in a vacuum of ? 10-4 mm Hg. In order to incorporate the maximum
amounts of these compounds into the polystyrene, and to study the conditions for thermal polymerization, the tem-
perature dependences of their solubilities in pure monostyrene were roughly determined. The results are given in
Table 1.
To avoid polymerization of the specimens during the solubility determinations (especially at the higher tem-
peratures), the experiments were performed rapidly (within 5 min) in a previously temperature-stabilized glycerine
bath.
1261
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 1. Solubilities of Phenyl Compounds in Styrene
at Various Temperature
Substance
M. p., ?C
Temp.,
?C
Solubility
in styrene,
wt. 10
Content
of ele-
mentary
metal,
wt. c/o
Pb(C6H5)4
224?
20
0,5
0,2
225
140
46
18,4
Bi(C6115)3
76
20
40
19,0
60
60
28,5
Hg(C6I15)2
120
20
20
11,3
60
60
33,9
As(C61-15)3
57
20
30
7,35
50
50
12,2
Sn(C6115)4
222--
20
0,5
0,14
223
140
.70
19,5
Se(C6I-15)2
Liquid
20
50
21,55
TABLE 2. Conditions for Polymerizing Styrene with Organometallic Additives
Substance
Content of
organometal-
lic compound,
wt. do
Content of ele-
mentary metal
in polystyrene,
wt. 070
Polymeriza-
tion tern-
perature, ?C
Polymeriza-
tion time, h
Properties
Pb(C6H5)4
Bi(C6H5)3
Hg(C6H5)2
As(C6H5)3
Sn(C6H5)4
Se(C61-15)2
30
40
60
50
40
30
12
19
33.9
12
11.1
10.2
140
160
120
120
120
150
140
16
144
60
144
24
170
Transparent, hard
Ditto, with yellow tint
Transparent, hard
Ditto
Transparent, hard,
rubbery
Preliminary experiments showed that, in the presence of these hetero-organic compounds, polymerization
cannot be effected by the procedure developed in [8] for preparing polystyrene-based PS. Each compound re-
quires definite temperature conditions for polymerization. For this reason the polymerization conditions were
investigated for each compound separately.
Polymerization Conditions for Styrene Containing Tetraphenyl Lead. Experiments
on the incorporation of tetraphenyl lead in styrene showed that, if thermal polymerization is carried out at 200?C,
addition of 10, 15, or 20 wt. 07o tetraphenyl lead causes blackening of the product. With 10 wt. /0 tetraphenyl lead
blacking occurred 6 h, and with 20 wt. % 4 h after polymerization began. In every case blackening began in the
lower part of the test flask, and during polymerization gradually penetrated the upper part of the specimen. It
should be noted that in all cases the specimen became yellow before turning black. This may be due to the fact
that the Pb(C6H5)4 molecule (m. p., 224-225?C) is markedly dissociated at these temperatures owing to the radical
processes involved in styrene polymerization. In fact, on decreasing the polymerization temperature to 190, 180,
or 170?C.the blackening was lessened.
The conditions for thermal polymerization of styrene containing Pb(C6H5)4 are as follows: a sealed ampoule
containing pure styrene and the required amount of tetraphenyl lead is placed in a glycerine bath at 145?C. The
contents of the ampoule are kept under nitrogen. The bath is gradually heated over a period of 6 h to 160?C, then
1262
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 3. Light Outputs of PS with Organometallic
Additives
Additive
Rel. output on excitation
Composition
Wt. %
y -rays
UV light
(2967 A)
Pb(C6H5)4
{10
Hg(C61-15)2 {
Sn(C6H5)4 {
11
5
10
100
75
54
50
30
95
79
100
90
85
80
64
100
100
kept at this temperature for a further 10 h. With this procedure, 30% tetraphenyl lead was added to the polystyrene
(12 wt. 10 elementary lead), To avoid turbidity, the products were cooled by a stream of water immediately on
completion of polymerization. Specimens prepared in this way were transparent and possessed adequate mechanical
strength. They do not alter on being left in the light, and are stable in air.
Polymerization Conditions for Styrene Containing Tetraphenyl Tin. Products con-
taining large quantities of tetraphenyl tin do not become blackened during thermal polymerization at 200?C. Sud-
den temperature changes during polymerization involve no risk, as the possible precipitation of a crystalline tin
compound on temperature reduction can be eliminated by quickly raising the bath temperature?no visible changes,
such as becoming black or yellow, then being observed.
The conditions for polymerization of styrene containing Sn(C6H5)4 are as follows: a sealed ampoule, con-
taining styrene plus the required amount of tin compound in an atmosphere of nitrogen, is placed in a bath pre-
viously heated to 135?C. The temperature is smoothly raised to 145-150?C over a period of 4 h. At this tempera-
ture polymerization goes to completion in 24 h. The specimens must be cooled like those containing tetraphenyl
lead. The amount of tin added is 11,1 wt. 50. The product is transparent, colorless, and stable in light and air.
Polymerization Conditions for Styrene Containing Diphenyl Mercury, Triphenyl
Bismuth, Triphenyl Arsenic and Diphenyl Selenium. Preliminary experiments on the incorpora-
tion of triphenyl bismuth in polystyrene showed that, with polymerization at 200?C, blackening occurs as for tetra-
phenyl lead. In the same polymerication conditions specimens with diphenyl mercury and diphenyl selenium turned
yellow, and the former became brittle. The specimens with triphenyl arsenic were colorless but brittle. The brit-
tleness sharply increased with increasing concentrations of the phenyl compounds of Hg, Bi, As, and Se (20, 30, 40,
and 50 wt. %). To eliminate this undesirable property, especially at maximum concentrations [60% Hg(C6H5)2,
40% Bi(C6H5)3, 50% As(C6H5)3], polymerization should be performed at 120?C. With 30 wt. %diphenyl selenium
the optimum polymerization temperature is 140?C. The polymerization conditiohs for various additives are com-
pared in Table 2. The products containing these elements are adequately hard and transparent. Despite prelimi-
nary purification of the initial materials, the specimens with triphenyl bismuth all had a pale yellow tint. The
polymers with diphenyl selenium, though transparent, were rubbery and had a reddish tint owing to the presence
of colloidal selenium in the diphenyl selenium. The styrene containing additives was polymerized in the thermo-
stat under nitrogen in sealed glass ampoules.
Effect of Organometallic Additives on the Luminescence of PS
To prepare scintillators, organometallic compounds were added to the styrene together with luminescent
substances-3 wt. %terphenyl and, as spectral displacer, 0.08% 1,3,5-triphenyl-2-pyrazoline. Polymerization of
the specimens with additives was performed in the conditions given in Table 2, and for the pure specimens in the
usual conditions [8]. Cylindrical specimens, of diameter 29 mm and length 26 mm, were cut from the polymeri-
zation products. A photomultiplier and mirror galvanometer were used to compare the light intensities from the
1263
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
TABLE 4, Luminescence of PS with p-Terphenyl
and Diphenyl Mercury
Concentration
of Hg(C6H5)2,
wt. lo
18
If
r, nsec
0
100
100
100
2.2
1
90
99
250
2.2
2
73
92
290
2.1
3
54
85
310
2.0
4
48
82
340
1.9
5
41
78
330
1.9
6
33
75
280
1.9
Note. Here I8 = luminescent intensity on excitation
by 5-rays from Gel"- Pr144; If = ditto on excita-
tion by UV light (X = 3130 A); IR = ditto on exci-
tation by x-rays from 60 kV tube; r = mesn dura-
tion of fluorescence on excitation by x-rays. PS
diameter 16 mm, length 10 mm.
quenched
specimens on excitation with y -rays from Co137. Teflon
reflectors were used to collect the light. The results
(Table 3) agree with the data for diphenyl mercury [12]
and for pressed scintillators [13].
From Table 3 it is evident that the presence of
organometallic compounds quenches the luminescence
of PS. It was of interest to discover at what stage of the
scintillation process this quenching occurs. A phase
fluorometer was used to measure the duration of lumine-
scence of the scintillators on direct excitation of the
spectrum displacer with UV light (X ks 3600 A). This was
found to be independent of the presence of the organome-
tallic compounds, i. e., the fluorescence of the displacer
is not quenched. Measurements of the fluorescent yield
of p-terphenyl on direct excitation in PS by the line
X ks 2967 A revealed the occurrence of quenching in the
presence of tetraphenyl lead, and especially of diphenyl
mercury (see Table 3). However, the degree of quenching
was much less than with excitation by y -rays. Quench-
ing by diphenyl mercury was also studied in a scintillator
with a single luminescent additive (1.5% p-terpheny1).
From Table 4 it is evident that radioluminescence is
less strongly than fluorescence of terphenyl. So the quenching of radioluminescence occurs mainly during
transfer of excitation energy from the polystyrene to the p-terphenyl.
PS containing organometallic compounds might be useful for measuring soft x-rays. For example, the absorp-
tion of bremsstrahlung x-radiation (u = 60 kV) by PS increases by a factor of eight on adding 5% diphenyl mercury.
Despite the reduced yield, the fluorescent intensity is increased more than three times (see Table 4).
LITERATURE CITED
1, H. Kallman, M. Furst, and F. Brown, Nucleonics, 14, No. 4, 48 (1956).
2. A. Ronzio, Internat. J. Appl. Radiation Isotopes, 4, 196 (1959).
3. R. Axtmann and Le Conter Cathey, Internat. J. Appl. Radiation Isotopes, 4, 261 (1959).
4, H. Gilman, E. Weipert, and F. Hayes, J. Organ. Chem., 23, 361, 628, 760, 910 (1958).
5. J. Birkes, Scintillation Counters [Russian translation], Moscow, Izd-vo, Inostr. lit. (1955).
6. L. Pichat and Y. Kochlin, J. Chim. Phys., 48, 225 (1951).
7. W. Buck and R. Swank, Nucleonics, 11, No. 11, 48 (1953).
8. E. A. Andreeshchev et al., Pribory i tekhnika eksperimenta, No. 1, 32 (1956); Izv. AN SSSR, Ser. Fiz., 22,
67 (1958); Zh. Fiz. Khimii, 34, 665 (1960).
9. E. E. Baroni and V. M. Shoniya, Atomnaya nergiya, 6, 330 (1959).
10. L. Pichat, P. Pesteil, and J. Clement, I. Chim. Phys., 50, 26 (1953).
11. M. Hyman and J. Ryan, IRE Trans. Nucl, Sci., MS-5, 87 (1958).
12. L. Basile, J. Chem. Phys., 27, 801 (1957).
13. E. A. Andreeshchev et al., Pribory i tekhnika 6ksperimenta, No. 4, 151 (1961).
14, A. Restaino, Nucleonics, 15, No. 9, 188 (1957).
15. N. Hosgen and R. Sowden, Nucleonics, 19, 158 (1961),
16. E. E. Baroni, T. N. Lebsadze, and V. M. Shoniya, Author's Certificate 115, 458 (March 11, 1958).
17. H. Gilman and J. Robinson, J. Amer. Chem. Soc., 51, 3112 (1929).
18. P. Preiffer and P. Truskier, Ber. Dtsch. Chem. Ges., 37, 1127 (1904); P. Preiffer and H. Pietsch, Ber. Dtsch.
Chem, Ges., 37, 4622 (1904).
19. Organic Syntheses [Russian translation], Symposium 2, Moscow, Izd-vo. Inostr. lit. (1949), p. 237.
1264
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
IN MEMORY OF KONSTANTIN KONSTANTINOVICH AGLINTSEV
from his fellow workers
Translated from Atomnaya gnergiya, Vol. 17, No. 6,
p. 501, December, 1964
Professor Konstantin Konstantinovich Aglintsev, Doctor of Technical Sciences, the head of a department and
laboratory at the V. G. Khlopin Radium Institute where he was one of the leading scientists, has died after a long
illness.
His scientific activity was entirely devoted to problems-of dosimetry, radiometry and the measurement of
ionizing radiation. His work was closely linked with the protection of the people's health. As an outstanding spe-
cialist in nuclear physics, he enjoyed considerable authority among Soviet and foreign biophysicists and doctors.
Aglinstsev was born in Leningrad in 1905, and after graduating from the Physics and Mathematics Faculty of
Leningrad State University he began work in 1926 in the X-Ray and Radiological Institute; from 1928 onwards his
career was for many years linked with VNIIM, the D. I. Mendeleev All-Union Research Institute of Metrology. His
early work was on x-rays, in particlar on their standardization. He continued to study these problems to the end of
his life. In 1940 he became head of the VNIIM radiometric laboratory and carried out important investigations on
metrology and radioactivity, aimed in particular at setting up a USSR state standard of radiation and the develop-
ment of new absolute methods for measuring radioactivity. During this period he worked on the national coordination
of measurement and the rational organization of a State measurement and instrumentation service for ionizing
radiation.
In 1950 he published the first edition of "Dosimetry of Ionizing Radiations": this is his chief publication and
at once became the principal text in the field. It was later revised by the author. Translations of this important
work were soon issued in Poland? Czechoslovakia and East Germany. Aglintsev wrote eight books and his total publi-
cations numbered eighty-two; much of his work is also to be found in the archives of scientific institutions.
His investigations of radioactivity naturally led him to the Radium Institute, where in 1950 he carried out work
the new field of spectroscopy by a and y-ray dosimetry. He devised a means for the quantitative detection of y-rays
by means of ionization chambers, counter tubes and photographic plates.
He established the shapes of the effective electron spectra in the fields
of various 6-emitters. He discovered a simple, important empirical law
on the constancy of the dose per 6-particle for a given 6 spectrum.
Throughout his life Aglintsev continually worked in scientific
education. He lectured at Leningrad State University (where many of
us were taught by him) and also in the Mechanical Institute, where he
was head of a department. In his work at the Radium Institute he paid
much attention to the education of his fellow-workers?not only his
immediate associates, but also those in other laboratories. His students
include many who are now working on lines originally laid down in
his work.
He has repeatedly represented Soviet science in the international
forum; even in the last few years he attended eight conferences and
symposia abroad. In 1959 at the Ninth International Radiological Con-
gress he was elected as a member of the International Commission on
Radiological Units and Measurement.
Besides this strenuous scientific, technical and teaching ability
Aglintsev was always active in social work. His responsiveness and
1265
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
attention to social interests won him the sympathy and recognition of his associates: he was elected to the
Leningrad Town Council of Workers Deputies.
His activities won him a number of Government awards. He was twice decorated with the Order of Lenin,
twice with the Order of Honor, and received the Leningrad Defense Medal, and the medal "For Valorous Work in the
Great Patriotic War."
Aglintsev's premature death has robbed us of a man of great soul, who loved life and people. He had friends
in very varied spheres and was also at home in the world of sport.
All will remember his great sympathy and his attentiveness to the interests and needs of his associates, espe-
cially the young. They came to him for advice, support, instruction and ideas, not only as to a long-standing scien-
tific associate, but also as to a good friend, always ready to give aid in word and deed.
We will always remember Konstantin Konstantinovich Aglintsev as a scientist, a man and a friend.
1266
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
LETTERS TO THE EDITOR
STUDY OF THE TRANSFORMATIONS OF RUTHENIUM DIOXIDE
IN THE PRESENCE OF CHROMIUM OXIDE
(UDC 546.96)
M. K. Baranaev, V. G. Vereskunov, and K. P. Zakharova
Translated from Atomnaya Energiya, Vol. 17, No. 6,
pp. 502-503, December, 1964
Original article submitted Mary 16, 1964
A study of the conditions for the formation of Ru04 is necessary in view of the considerable amount which
passes into the vapor phase during the heat treatment of atomic industry waste. Usually either nitric acid solutions
or hydroxide pulps are subjected to heat treatment; these contain a mixture of fission products, including Rul".
Apart from technological impurities, such solutions and pulps contain corrosion products, especially chromium
and iron.
We investigated transformations taking place on heating ruthenium hydroxide and a mixture of this with chro-
mium hydroxide. The ruthenium hydroxide used was obtained by alkali precipitation of ruthenium choride. The
hydroxide precipitates were washed to remove all traces of alkali and chlorine ions, and then dried at a temperature
of 25 to 27?C.
Determination of the volatilization of the Ru04 was made with ruthenium hydroxide obtained by alkali preci-
pitation of a tagged solution of RuC13. The radioisotope of ruthenium was introduced in the form of the chloride of
Rul". The chromium hydroxide was obtained by alkali precipitation of a solution of chromium nitrate. The behav-
ior of the material under examination was studied by the complex thermal analysis method [1]. The transformation
of the ruthenium into the vapor phase during heating, associated, as shown in [2, 3] , with the formation of volatile
Ru04, was estimated from the change in the activity of the sample, measured on an end-window counter MST-17.
-In order to reduce the effect of self-diffusion, we used samples with a material layer thickness not exceeding
100 mg/cm2. The thermal analysis was made with large quantities of a stable ruthenium isotope.
As a result of thermal analysis of the hydrate of ruthenium oxide, it was established that the loss of water from
the material took place in stages in three temperature ranges (endothermic effects at 100, 260, and 330?C) and
ceased at 360?C. In this process, as shown earlier [2], RuO2 was formed. The loss of weight* from the ruthenium
hydroxide in the temperature range indicated was 15%. Theoretically, passage from ruthenium hydroxide, bearing
the formula Ru205 2H20 [2], to RuO2 should involve a 16% loss in weight. The weight loss of the sample at tempera-
tures above 700?C is associated with the oxidation of RuO2 to volatile Ru04 beginning at these temperatures [2, 3].
The exothermic effect observed around 410?C corresponds to the transformation of RuO2 from the amorphous
into the crystalline state, as confirmed by x-ray analysis. The crystalline phase is absent from unroasted ruthenium
hydroxide. X-ray diffraction patterns of samples roasted at 475?C show sharp lines indicating the presence of a
crystalline phase.
In the case of chromium hydroxide, the exothermic effect associated with the transformation of chromium
oxide from the amorphous into the crystalline state takes place at 450?C. In the thermal diagram of the binary
mixture Ru203. 2H2O?Cr(OH)3, however, no exothermic effects appear, and this strongly suggests interaction between
the substances in question.
In order to confirm this, we studied the volatilization of ruthenium on heating a mixture of tagged ruthenium
hydroxide and chromium hydroxide in air and in nitrogen. The sample spent 30 min in each temperature range.
Volatilization of ruthenium on heating the mixture began around 400?C. It was shown in [2, 3] and confirmed
by our own experiments that marked oxidation of RuO2 by atmospheric oxygen to volatile Ru04 only took place at
*Determined by weighing the samples.
1267
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
temperatures above 700?C. Since no ruthenium evaporated on heating the mixture to 900?C in nitrogen, we may
assume that in the present case chromium oxide does not constitute an oxidizing agent, but simply catalyzes the
process, leading to the oxidation of Ru02 at lower temperatures. The catalytic power of chromium oxide, associated
with its transformation into a form containing active oxygen on heating in air, is examined in detail in [4].
The insignificant volatilization of ruthenium on heating the mixture in a nitrogen atmosphere is determined by
the reaction of disproportiOnization of Ru02 [2]. As a result of this reaction, which takes place above 800?C, Ru04
and Metallic ruthenium are formed.
Our experiments thus show that, on heating the hydrate of ruthenium oxide to a temperature around 410?C,
there is an exothermic effect corresponding to the transformation of Ru02 from the amorphous into the crystalline
state. They also show the catalytic effect of chromium oxide on the oxidation of Ru02 by atmospheric oxygen to
Rt104 at 300-700?C.
LITERATURE CITED
1, L. G. Berg and G. G. TsurinOV, The Kurnakov Pyrometer [in Russian], Moscow, izd, AN SSSR (1942).
2. P. Pascal, Nouveau traite de chimie minerale, 19, Paris (1958).
3. L. Grnelin, Grnelins Handbuch der Anorganischen Chemie, H, 6.3, Berlin (1926).
4. T. V. Rode. Oxygen Compounds of Chromium and Chromium Catalysts [in Russian], Moscow, izd.
AN 8SSR (1962).
1268
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
USE OF THE TIME OF FLIGHT METHOD FOR MEASURING
THE RANGE/ENERGY RELATION FOR 18 TO 38 MeV
HELIUM IONS IN ALUMINUM
(UDC 539.128.53)
N. I. Venikoy and N. I. Chumakov
Translated from Atomnaya Energiya, Vol. 17, No. 6,
pp. 503 -504, December, 1964
Original article submitted November 20, 1963
The time of flight method [1] is suitable for measuring the energies of ions absolutely to a high degree of
accuracy. We used this method in determining the range/energy relationship for He3 ions in aluminum over an
energy range of 18 to 38 MeV. The measurements were made in the emergent beam of the 1.5-m cyclotron in the
I. V. Kurchatov Institute of Atomic Energy [21, using a multichannel time analyzer described in [3]. Around 10% of
the ion beam deflector by the sectoral magnet fell on a graphite target, and the remaining ions were intercepted by
a graphite grid set at a distance of 5515 mm in front of the target. The y-quanta from the target and grid were
recorded simultaneously by the time analyzer. In order to calibrate the time scale, recordings were made of y-peaks
from two high frequency periods. So that the intensity of y-peaks from the target and grid should be approximately
the same, the y-quantum detector, a stilbene crystal with an FEU-36 type photomultiplier, was placed at a distance
of roughly 2.8 m beyond the target. In order to eliminate this distance from the calculation, the detector was set
at zero angle to the beam axis. Use of short ion bursts (less than 2 nsec) ensured an accuracy of around 0.15 nsec in
the accuracy of time of flight measurements by the analyzer.
As we know, the kinetic energy of the ion is
Here E0 is the ion rest energy,
E=_-E0 ( 1 - i)
171-132
where y is the velocity of the ion, and c is the velocity of light.
If we confine ourselves to two terms in the expansion of expression (1), we obtain
in which
13= T 10-6 L , '
f c )
-I-
2T 3 -\
(1)
(2)
(3)
where L is the distance from grid to target in cm, T is the time of flight of an ion from the grid to the target, T is
the time interval between two y-peaks from the target (or grid), corresponding to the high frequency period, and f is
the frequency of the cyclotron hf generator in Mc/sec. The error in measuring the energy of the ions is given by
the expression
AE AL , At AT , Ari
-E =21 7,?r--1-717-7-f_l'
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
(4)
1269
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Range/Energy Relationship for He3 Ions in Aluminum
Energy, MeV
18,15
22,29
24,34
27,01
30,34
32,82
35,95
37,60
' Measured range, mg/cm2 . . .
53,1
74,2
86,9
104,4
128,9
147,3
162,7
190,6
Range as given in [4], -
mg/cm2
54,3
76,5
89,4
107
?
?
?
?
In our case, L was measured with an accuracy of 2 mm, AT and AT were around 0.15 nsec for T = 85 to
120 nsec, and T = 120 to 165 nsec, while Lf = 5 kc/sec for f = 8.3 to 12 Mc/sec.
In all the maximum error in determining the ion energies was no greater than ?0.4%.
In order to determine the range of the ions in aluminum, a pile of foils was placed at intervals in the path of
the beam, the total foil thickness being so chosen as to reduce the intensity at the target by a factor of two. To
eliminate error arising from secondary emission, the target was made in the form of a Faraday cylinder and placed
in a magnetic field (? 200 Oe) directed along the normal to the cylinder axis. In order to control the stability of the
beam falling on the foils, a current monitor in the form of a grid was placed in front of these. Piles of foils were
turned on a lathe (the diameter of each foil being 60 mm) to an accuracy of 0.06 mm, and then weighed to an ac-
curacy of 1 mg. The error in determining the range may be found from the formula
AR Ap AD cp2
L 2- ;
. p D 2
(5)
p i
where - - is the weighing error, equal to 2- 10 , n determining the diameter of the pile of foils, equal
to 10-3, cp is the maximum deviation of the direction of the beam incident on the foil from the normal, equal to
0.05 (31, and g is the error due to foil inhomogeneity and some current instability, less than 2- 10-3. The over-all
error in determining the range is no more than ?0.3%.
The table shows the range/energy relationship measured for He3 ions in aluminum; for comparison, some range
values taken from the computed curve of [4] are presented. No other data on the range/energy relation for He3 in
aluminum in the energy range considered are known to the authors. As seen from the table, the values given in [4]
for the range of He3 are a little (2.5 to 4%) higher for comparable energies. The precision of measuring the range/
energy relation by the method described is also confirmed by the good agreement (better than 1%) between the
range/energy relation measured by the authors using the same method for 6 to 17 MeV protons in aluminum (12 points
in all) and that given in [4].
The authors are grateful to N. A. Vlasov for valuable comments, to S. P. Kalinin, V. P. Rudakov, and
R. V. Rybakov for interest in the work, and also to V. D. Krupochkin, V. V. Paramonov, and B. I. Khoroshavin for
assistance rendered.
LITERATURE CITED
1. B. V. Rybakov and V. A. Sidorov, Spectroscopy of Fast Neutrons [in Russian]. Supplement No. 1 to the
magazine "Atomnaya Energiya," Moscow, Atomizdat (1958).
2. A. A. Kurashov and V. A. Sidorov, "Pribory i tekhnika eksperimenta," No. 6, 69 (1961).
3. D. Bromley and E. Almigvist, Reports on Progress in Physics, AECL, No. 950, Chalk River, Ontario (1959).
4. Experimental Nuclear Physics, Editor E. Segre, 1, [Russian translation], Moscow, IL, (1955), p. 190.
1270
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
CHARGE-EXCHANGE OF OXYGEN IONS OF ENERGY 2-13.3 MeV
IN THIN ALUNDUM FILMS
(UDC 539.188)
N. I. Venikov, N. I Chumakov and 3. I. Khoroshchavin
Translated from Atomnaya tnergiya, Vol. 17, No. 6,
pp. 504-505, December, 1964
Original article submitted November 20, 1963
In connection with linear accelerators for oxygen-ions acceleration and electrostatic generators with charge-
exchange, interest attaches to the relation between the concentrations of variously charged ions and the energy of
the "stripped" beam. A relation of this type was obtained in experiments, performed with the cyclotron of the
I. V. Kurchatov Institute of Atomic Energy, in which Oa ions were accelerated with the third, fifth and seventh hf
subharmonics. The ion beam from the cyclotron impinged on an alundum film of thickness 20 pg/cm2 and area
20 x 40 mm2, prepared by a method similar to that in [1]. After passing through the film the ion beam was analyzed
according to charge by a sector magnet. To eliminate errors due to secondary emission from the target, the ion de-
tector (a conventional Faraday cylinder) was mounted in a magnetic field of intensity ? 100 Oe. The energies of
the ions penetrating the alundum film were determined
from the field of the sector magnet, which was previously
calibrated to within ?1% throughout its range, using a
0, beam of protons whose energy was determined from their
time of flight.
UI
1111?
1M
1E111111011
'-3/13101111111111
nail
III
111,111=11
MI
PlillitiLl?
0,4
0
0
41
o ?f 4
6 8 10 f2 E, MeV
Fig. 1. Relative concentration nz of variously charged
016 ion, plotted versus energy of beam penetrating
film.
3
4
3
0 2
4
8 10 E, MeV
Fig. 2. Equilibrium charge of 016 ions versus energy.
Fig. 1 plots the relative concentrations of variously
charged 016 ions versus the energy of the beam penetrating
the alundum film.
In calculating the angular scattering of a beam on
passing through a thin tharget, it is necessary to know the
equilibrium charge on the ions. Fig. 2 plots the energy de-
pendence of the equilibrium charge on the 016 ions, calcu-
lated as
where 7. is the equilibrium charge, ni the concentration of
ions with charge ; (in our case Zni I).
The authors would like to thank A. A. Shubin for
help in preparing the alundum film.
LITERATURE CITED
1. G. B. Andreev et al., Pribory i tekhnika eksperimenta,
No. 6, 149 (1961).
1271
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
Declassified and Approved For Release 2013/03/14: CIA-RDP10-02196R000600120002-4
NEUTRON RADIATIVE CAPTURE IN COPPER AND MOLYBDENUM
(UDC 539.172.4)
V. A. Tolskikov, V. E. Kolesov, A. G. Dovbenko,
and Yu. Ya. Stavisskii
Translated from Atomnaya gnergiya, Vol. 17, No. 6,
pp. 505-508, December, 1964
Original article submitted October 10, 1963
The energy dependence of the radiative capture cross section in the isotopes Cu65 and Mol?? was measured ex-
perimentally for nuetrons with energies of 5-200 keV. The measurements were made using the relative activation
method in a ring geometry. The reaction Li(p, n), produced in a Van de Graaf accelerator, was used as the neutron
source. The technique of measurement has been described in [1, a
Normalization of the Cu65 cross sections was done by using a weighted mean value of 44.9 ? 1.7 mb for the
radiative capture cross section of neutrons from an Sb?Be source [1, 3,4]; the neutron energy was 24 keV. The
6
100
70
1
IN=NriallE.
111?1111.111MMINIIII
w
MI NM
um
millIMINIMENN
_i_
MIMI
mum
1?11111111111111111111111
MI
mimi
IIIHIII 11111
?-11
III
1.,0 ?
Ill
I
I
-