SOVIET ATOMIC ENERGY VOL. 59, NO. 6

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP10-02196R000300070006-9
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 103
• Document Creation Date: December 27, 2016
• Document Release Date: February 20, 2013
• Sequence Number: 6
• Publication Date: December 1, 1985
• Content Type: REPORT

Declassified and Approved For Release 2013/02/20 :CIA-RDP10-021968000300070006-9 '
Russian Original Vol. 59, No. 6, December, 1985
June, 1986
SATEAZ 59(6) 957-1054 (1985)
SOVIET
ATOMIC
ENERGY
ATOMHAA 3HEPfNA
(ATOMNAYA ENERGIYA)
U
TRANSLATED FROM RUSSIAN
CONSULTANTS BUREAU, NEW YORK
Declassified and Approved For Release 2013/02/20 :CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20 :CIA-RDP10-021968000300070006-9
,~ ~ U V 1 E T - I ~ f ~uvrer H comic energy Is a translation of Atomnaya Energiya, a
` publication of the Academy of Sciences of the USSR.
ATOMIC
ENERGY
Soviet Atomic Energy is abstracted or~in-
dexed in Chemical Abstracts, Chemical
Titles, Pollution Abstracts, Science Re-
search Abstracts, Parts A and B, Safety
Science Abstracts Journal, Current Con-
tents, Energy Research Abstracts, and
.Engineering Index- _
Mailed in the USA by Publications
Expediting, Inc., 200 Meacham Ave-
nue, Elmont, NY 11003.
POSTMASTER: Send address changes to
Soviet Atomic Energy, Plenum Publish;
~ing Corporation, 233 Spring Street, New_
York, NY 1001 3.
An agreement with the Copyright Agency of the USSR (VAAP)~'
makes available both advance copies of the Russian journal and
original glossy photographs and artwork. This serves to decrease
the necessary time lag between publication of the original, and
publication of the translation and helps to improve the quality
of the latter. The translation began with- the fi: st issue of the- ,
Russian journal. ~ - ~ , -
Editorial Board of Atomnaya ~nergiya:
Editor: 0. D. Kazachkovskii
Associate Editors: A. I. Artemov, N. N. Ponomarev-Stepnoi,
-and N. A. Vlasov
.,
1. A. Arkhangel'skii
I. V. Chuvilo
I. Ya. Emel'yanov ,
I. N. Golovin -
V.I.II'ichev. -
P. L. Kirillov
Yu. I. Koryakin ?
E. V. Kulov
B. N. Laskorin
V. V. Matveev
A. M. Petras'yants
E. P. Ryazantsev
A. S: Shtan
B. A.,Sidorenko
Yu. V. Sivintsev
M. F: Troyano -
V. A. Tsykanov
E. I. Vorob'ev
- V. F. Zelenskii
Copyright ?1986, Plenum Publishing Corporation. Soviet Atomic Energy partici-
pates in the Copyright Clearance Center ICCCI Transactional Reporting Service.
The appearance of a code line at the bottom of the first page of an article in this
journal indicates the copyright owner's consent that copies of the article maybe
made for personal or iriternat use. However, this consent is given on the condition
that the copier pay the flat fee of S9.50 per article Ino additional per-page fees!
directly to the Copyright Clearance Center, Inc.,. 27 Congress Street, Salem, ,
Massachusetts 01970, for all copying not explicitly permitted by Sections 107 or
108 of the U:S. Copyright Law. The CCC is a nonprofit clearinghouse for the pay-
ment of~photocopying fees by libraries and other users registered with the CCC.
Therefore, this consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new
collective works, or for resale, nor to the reprinting of figures, tables, and text ex-
cerpt`s. 0038-521 X/85/509.50 ,
Consultants Bureau journals appear about six months after the publication of the
original Russian issue. For bibliographic accuracy, the English issue published'by
-Consultants Bureau carves the same number and date as the original Russian from
which it was translated. For example, a Russian issue published~in December will
appear in a Consultants Bureau English translation about the following June, but the
translation issue will carry the December date. When ordering any volume or particu-
lar issue of a Consultants Bureau lournal,.please specify the date and, where appli-
cable, the volume and issue numbers of the original Russian.-The material you will
receive will be a translation of that Russian volume or issue.
Subscription (2 volumes per year)
Vols. 58 & 59: 5645 (domestic!; 5715 (foreign)
Vols. 60 & 61: S695 (domestic); S770 (foreign) Single Article: 59.50
CONSCIITANTS BUREAU, NEW YORK~AND LONDON
6
0
233 Spring Street
New York, New York 10013
Declassified and Approved For Release 2013/02/20 :CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
A translation of Atomnaya Energiya
Volume 59, Number 6 17ecember, 1985
C?WTENT8
Engl./Russ.
ARTICLES
Physicochemical Foundations of the Modeling of the Composition
of the Water Coolant in a Nuclear Power Plant - V. M. Sedov,
L. V. Puchkov, V. G. Kritskii, and V. I. Zarembo. . . . .
Problems of~Chemical-Analytical Monitoring in Nuclear Power
- L. N. Moskvin . . . o . . . .
Removal of Corrosion Products from the Steel Surfaces in the Aqueous
Coolant of Nuclear Power Plants -? V. G. Kritskii, A. S. Korolev,
I. G. Berezina, and M. V. Sof'in. . . . o . . . . .
Calculation of the Magnitudes of Deposition and Concentration
of Corrosion Products in Boiling Water Reactors - 0. T. Konovalova,
M. I. Ryabov, L. N. Karakhan'yan, and T. I. Kosheleva . . .
Formation of Deposits on the Surface of the Fuel Elements of RBMK-1000
- I. A. Varovin, S. A. Nikiforov, A. P. Eperin, Yu. N. Aniskin,
V. G, Kritskii, and Yu. A. Khitrov. . . . . . . .
Reasons for and Against the Oxygen Dosage in Condensate Feed Circuits
of Nuclear Power Plants with RBMK-1000 Reactors - V. V. Gerasimov,
A. I. Gromova, V. N. Baranov, and Yu. V. Makarenkov . . . . . . .
Study and Selection of New Extractants for Actinide Extraction
- A. M. Rozen, A. S. Nikiforov, Z. I. Nikolotova,
and N. A. Kartesheva . .
Mathematical Model. of the Temperature Fie]:d?around a Borehole
with Radioactive Wastes and Its Experimental Verification in Field
Conditions - E. G. Drozhko, V. I. Karpov, Ao. S. Stepanov,
I. I. Kryukov, V. F. Savel'ev, V. V, Kulichenko, V. A. Bel'tyukov,
and A. A. Konstantinovich . . . . . . . . .
Passage of Primary Protons through a Shield with a Random Distribution
of the Material - V. G. Mitrikas, V. M. Sakharov,
and V. G. Semenov . .. . .
Measurement of the Neutron-Induced Fission Cross Sectior. Ratio s?
of z3fiU and z3sU for Energies of 4-11 MeV - A. A. (roverdovskii,
A. K. Gordyushin, B. D. Kuz'minov, A. I. Sergachev,
V. F. Mitrofanov, S. M. Solov'ev, and T. E. Kuz'mina. . .
Fields of Ionizing Radiations on the Tokamak-10 Fusion Unit
- Vo S. Zaveryaev, G. I. Britvich, V. I. Lebedev, Vo S. Lukanin,
F. Spurny, I. Potochkova, and I. Kharvat. .
Distribution of Lead in Rocks by the Method of Fission-Fragment
Radiography - V. P. Perelygin, G. Ya. Starodub,
and S. G. Stetsenko . . . . . . . .
957
395
962
398
965
401
968
403
971.
405
976
409
982
413
994
422
999
425
1004
429
1008
432
1015
437
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9 ;~g~
---- .
Calculation of Creep Contours of Textured Zirconium Alloys along Polar
Figures - S. B. Goryachev, A. V. Shalenkov, and P. F. Prasolov. .
Three-Dimensional Calculations of a Subcritical Heterogeneous Reactor
with a. Neutron Source - V. M. Malofeev
Influence of the Finite Moderator Dimensions upon the Characteristics
of a Pulsed Source of Slow Neutrons - N. I. Alekseev,
A. V. Drobinin, and Yu. M. Tsipenyuk. . .
Liquid Reference Sources of Gamma Radiation - B. Ya. Shcherbakov .
A Specialized Mass-Spectrometer Unit for Analyzing Aggressive Gas
Mixtures - N. N. Bobrov-Egorov, V. N. Ignatov, and G. I. Kir'yanov. .
Equivalent X-Ray Doses in a Heterogeneous Human Phantom - V. I. Ivanov,
L. A. Lebedev, V. P. Sidorin, R. V. Stavitskii, and V. V. Khvostov. .
INDEX
Author Index, Volumes 58-59, 1985.
Tables of Contents, Volumes 58-59, 1985. .
The Russian press date (podpisano k pechati) of this issue was 12/2/1985.
Publication therefore did not occur prior to this date, but must be assumed
to have taken place reasonably soon thereafter.
(confined)
Engl./Russ.
1019
439
1021
440
1024
442
1026
443
1028
444
1030
446
1035
1041
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 AR Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
V. M. Sedov, L. V. Puchkov, UDC 621.187:628.163.0015
V. G. Kritskii, and V. I. Zarembo
The operational reliability and safety of modern power plants are largely determined by
the water-chemical conditions. There now exist mathematical and physicochemical_models
which describe the interaction of the coolant with the structural materials and which take
into account the corrosion of iron. The models of the deposits of corrosion products on
equipment surfaces and the growth in the total radioactivity of the equipment. take into
account the fact that during mass transport the iron can be in the ionic, colloidal, and
undissolved (oxide) forms. These forms exist in virtually the entire temperature interval
of the coolant. Is is natural to expect that the mechanisms of deposition of these forms
on surfaces are different. Reducing the inflow of iron into the active zone of boiling
water reactors [1] has decreased the rate of growth of the total radioactivity of the equip-
ment. This could be linked to the fact that the solubility of iron in the coolant also
plays a determining role in mass transport of radioactive cobalt isotopes.
The diversity of the forms in which iron is present in the coolant poses the question
of the necessity of taking these forms into account in models of corrosion, mass transport,
and deposition. The possibility of developing such models is determined by the reliability
and accuracy of the data, on which the calculations are based, on the equilibrium solubility
of different forms of corrosion products as a function of the state parameters of the coolant,
the presence of different impurities and corrective additives in the coolant, as well as
dissolved gases.
The solution of these problems can be based on the methods of equilibrium thermody-
namics, which enable determining the number of different chemical forms of components, their
transformation with increasing temperature, pressure, concentration of correcting additives,
or dissolved gases, on the basis of a priori representations of the chemical composition
of the coolant, if the information required for the calculations on the thermodynamic func-
tions is available. Methods have now been developed for calculating the equilibria in multi-
component heterogeneous systems, and with the help of fast computers such problems are now
solved comparatively easily.
In this formulation the problem is one of obtaining reliable information on the standard
values of the Gibbs energy of formation of ions and charged or neutral ionic associates
in a water solution at high temperature and pressure. Two computational methods [2, 3]
are now primarily used to solve this problem; of these, the most widely used is the Criss-
Cobble "correspondence principle" [2], though in our opinion it has no advantages over
Khodakovskii's method [3]. This is apparently explained by the fact that foreign investi-
gators (and it is they who first began to use widely the methods of equilibrium thermo-
dynamics in order to analyze the interaction of structural materials with the coolant) pri-
marily use the correspondence principle.. Algorithms and programs which utilize this method
to calculate the Gibbs energy of different forms of dissolved components for T = 298-573?K
now exist [4].
Nevertheless the results obtained thus far do not satisfy investigators. The reason
for this lies in the fact that the above-indicated methods are limited to a definite tem-
perature interval. The Criss-Cobble correspondence principle was proposed by them for tem-
peratures < 473?K; in addition, they specifically stipulate that they are not responsible for
results obtained by extrapolation with the help of their method to higher temperatures.
Almost all investigators who use the correspondence principle in their calculations forget
this stipulation in practice. It should be noted that even the most highly active supporters
of this method do not present calculations for T > 573?K, though such calculations are un-
Translated from Atomnaya Energiya, Vol. 59, No. 6, pp. 395-398, December, 1985. Orig-
inal article submitted November 16, 1984.
0038-531X/85/5906-0957$09.50 p 1986 Plenum Publishing Corporation 957
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 do Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9 ~ of
bivalent cations in a water solution at T = 523-573?K computed with the help of this method
have a temperature coefficient whose sign is opposite to that obtained in experimental in-
vestigations. An altered variant of the correspondence principle for single-atom ions has
been extended up to 573?K [6]. This method is, however, limited to definite chemical forms
of the ions. Khodakovskii's method, which we proposed for T < 473?K, does not have .this
restriction. A comparison of the predictions and the experimental data shows that this
method can be used for singly charged ions at higher temperatures also (up to 573?K). How-
ever, the temperature extrapolation of the Gibbs energies with the use of this method must
be done with great care and the results obtained must be critically interpreted. Thus in
the calculation of the Gibbs energy of formation of NaRe04 ions, which have a positive limit-
ing partial molar heat capacity at 298?K, in the solution of a stoichiometric mixture, we
shall obtain a temperature dependence which is precisely opposite to that obtained experi-
mentally. This is valid for all stoichiometric mixtures of ions which have a positive or
close to zero limiting partial heat capacity.
The indicated methods have the common and principal disadvantage that they cannot be
used to calculate the Gibbs energy of formation of neutral ionic associates in solutions
at high temperature, especially since as the temperature is increased the molecular component
of the solubility in the overall concentration of the saturated solution increases [7].
For this reason, in the absence of experimental investigations of the dissociation constants
of electroneutral associates, the solubility at high temperature cannot be calculated thermo-
dynamically based only on the Criss-Cobble and Khodakovskii methods.
We propose a method for calculating the temperature and pressure dependences of the
Gibbs energy of formation of ions in a water solution whose correctness is based on compari-
son with experimental results on 26 binary water-salt systems, obtained in our laboratory
and taken from [8]. The method enables obtaining quantitative results for 298-873?K and
pressures from equilibrium pressure up to 500 MPa with water densities >0.4 g/cm3.
The equation for the calculation of the standard Gibbs energy of formation of an in-
dividual ion of any chemical form in solution at a temperature T and pressure p has the
form
ec?~ a' p=
s. q
X0,2?8 no-{- ~ V2i dp-(T-298) S{,298, To+.Cpr [(T-298)-
s, aq
Po
(1)
-T In 7'/298]-]-n~ {[GTao-GZ g -{-SZ g X
X(T-298)jvk+P-(Gazo-G2 8 -~-S2~s (T-298)~P)~-tlai (1/ET. P-1~E298. n)-{-~1Qij'zse, F~ (T-298),
where DG?..2es,p? and S?.'298'p0 are the standard Gibbs energy of formation of the ion in
the water's~lution andlits absolute entropy at 298.15?K and 0.1013 MPa, respectively; VZi,
absolute limiting partial molar volume of an ion in a water solution; Cpg, molar heat ca-
pacity of the ion in the perfect-gas state; ni, corodination number of the ion; GHs.O and
SHzO, Gibbs energy and entropy of formation of water, respectively; E, dielectric constant
of water; r~ = NAe2/8~rE? (NA is Avogadro's number; a is the electron charge; E? is the di-
electric constant); ai = zi/ri (zi is the ion charge and ri is the'ion radius, within which
the dielectric saturation of the solvent is admitted); Y298~p = 1/E(e In E/8T); pk, some
effective pressure, which has the same value for large groups of ions, determined by whether
they are cations or anions, single-atom or many-atom ions, as well as by their charge.
Let the dissociation occur according to the scheme
cC ~ aA-{- bB. ( 2 )
In this case, from the viewpoint of formal thermodynamics, the problem of the temperature
and pressure extrapolation of the electrolytic dissociation constants of a dissolved
charged ionic associate must be solved uniquely with the help of Eq. (1). -For the dissocia-
tion constnats of a dissolved charged ionic associate must be solved uniquely with the help
of Eq. (1). For the dissociation reaction HZP04- F H~" + HP04- the predictions coincide
practically completely with the experimental results [9] and the predictions of [10] (see
Fig. 1). In the case of the dissociation HC03- f H+ + C03- the difference from the data
in [10], obtained from an analysis of the experimental results of different authors, does
not exceed several pKdis? Based on the error in the data in [7, 10-12] the agreement of the
results may be regarded as satisfactory.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
?, .~._
/Hg(OH)==Hgz+20H
/ UOz(OHh=UOi+20H
HGOy= H++CO~
~~ o ? ? fHzP04=H`+HPO~
o--o--~~ o ? H=GO?y= H++HCO~
2 0
Y Q??? ? HSP04= H++H=P04
a 6 kgG6?o Ag` sGt_ z
4~d,, _ o~.~-~.Mg50y=My .SOy
300 400 500 600 700 T,I?K
Fig. 1. Comparison of pKdis for
several ionic associates, obtained
by a computational method, with the
published data: ~ -experimental
data [7]; ? -prediction [10], 0 -
data in [10] obtained based on an
analysis of experimental data 'o -
[11]; p -[12]; calculations of this
work.
TABLE 1. Standard Values of the Gibbs Energy of Formation of the Ionic Associate
NaCl? in a Water Solution at the Saturation Vapor Pressure of Pure Water,
kJ/mole
Temp., ?K
Source or method for
obtaining the value
373 I
423 I
473 I
523 I 573
388,02
394,38
400,13
406,62
413,22
418,54
-,17,41
(12]
388,02
393,1
396,8
400,8
405,1
409,4
414,1
Calc. with total dehydration
Calc. without dehydration
An important point in this case, from our viewpoint, is the assertion made in [13]
that in the case of dissociation according to the scheme (2), when the ions A and B combine
and form the charged ion C, the mechanism of uniform distribution of the total. charge of
the ions A and B over the sphere of the ion C is implausible. In the case of the formation
of a neutral ionic associate, however, it admits the compensation of the charges of the
A and B ions. The calculations presented show that the model determination of the dis-
sociation constant of a charged ionic associate by the scheme (2) is possible precisely
in accordance with the mechanism of uniform distribution of the total charge of the dis-
sociation reaction products over the sphere of the starting ion.
The calculation of the dissociation constants of charged ions was in practice prede-
termined by the proof of .the correctness of Eq. (1) [8], whereas in the case of the calcula-
tion of the Gibbs energy of formation of neutral ionic associates there arise three questions
whose solution will determine the possibility of their quantitative modeling.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
_- - -- --- ---------- ---- ----_,___ __ ~.r~ri-
mental studies performed in the monographs [7, 11] as well as attempts to examine the ques-
tion of the formation of ion pairs from the viewpoint of the electrostatic theory show that
in the first approximation this polarization can be neglected, i.e., in the formation of
a neutral molecule, more precisely, of a neutral ionic associate, the electric charge of
its constituent ions.is mutually compensated. This also follows logically from the con-
clusion drawn above regarding the uniform distribution of the total charge of the dissocia-
tion reaction products over the sphere of the starting ionic associate. Thus if it is assumed
that Eq. (1) is the general equation for calculating the Gibbs energy of formation of both
ions and ionic associates in a water solution, then in the case of the calculation of the
Gibbs energy of formation of a neutral ionic associate the last terms of this equation,
associated with the polarization hydration, vanish.
2. Does the specific hydration associated with the transfer of water molecules from
the solvent into the coordination sphere of the ions, forming an ionic pair, remain when
a neutral ionic associate forms, i.e., is the dehydration of ions accompanied by the forma-
tion of a neutral ionic asosciate? If so, then what is its significance?
3. What is the heat capacity Cp of the neutral asssociate?
The last two questions can be answered only by correct model calculations for extreme
cases. For the base system for answering the last two questions we selected the system
NaCl HZO, for which the Gibbs energy of formation of sodium and chlorine atoms as well as
the dissociation constant of the molecule NaCl? are known in a wide temperature interval
[12]. According to Eq. (1), from which the terms associated with polarization hydration
were eliminated, we calculated the values of the Gibbs energy of formation of the neutral
ionic associate NaCl with a saturation vapor pressure of the pure solvent for the case of
total hydration and for the case when the coordination numbers of the sodium and chlorine
ions are preserved by the ionic associate (see Table 1). These calculations presume that
Cpg (NaCl?) is equal to the maximum heat capacity of the NaCl molecule in the excited state
of the perfect gas (9/2) R. When the last factor is replaced by (7/2) R - the minimum heat
capacity of molecules in the unexcited state - we obtain a difference in the values of
aGf~agPo(NaCl?) not exceeding 1 kJ/mole at 623?K. Analysis of Table 1 shows that the predic=
tions in both cases are more positive than the experimental values. The predictions ob-
tained under the assumption of total dehydration, however, are closer to the experimental
values. Their difference does not exceed 3.3 kJ/mole at 623?K. Thus for calculating the
standard values of the Gibbs energy of formation of neutral associates in a water solution.
at the saturation vapor pressure of pure water the following particular case of Eq. (1) ,~
was obtained:
4Gf~ qP0=4Gt~ aq~~ Po-S;,2q8, Po (T-2qR) -f-C~g[(T-298)-T 1n T/298].
(3)
In the first approximation the problem of calculating the change in the Gibbs energy
of the reactions of different transitions with the participation of dissolved components
can be regarded as solved, of course, within the limits of error of present high-temperature
investigations. But, unfortunately, the method requires a knowledge of the thermodynamic
functions of the solution compoennts at 298.15?K and 0.1013 MPa, and this information is
widely available only for simple ionic forms; it is limited primarily to the Gibbs energy
for the charged ionic associates and there is virtually no information for neutral associates.
The method of quantitative evaluation of the entropy at 298.15?K is limited to the chemical
forms of the components of the solution, and quantities such as the partial volumes of the
neutral ionic associates in a water solution are not available at all.
We did not attempt to build another model of hydration. The need for such a model
followed from practical problems, primarily energetics. The purpose of such a model is
to establish the quantitative composition of complicated water-salt systems at high tempera-
ture and pressure on the basis of a priori information about their composition. Using the
required data at 298.15?K and 0.1013 MPa from [10, 13], the solubility of magnetite, goethite,
amakinite, FeO, and Fe(OH)3 in water for T < 623?K at the saturation vapor pressure of the
pure solvent has now been calculated by the method of minimization of the Gibbs energy.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
L11P~11HlU1tP~ 1.11L.L
1. Y. Mishima, "Study on the influence of water chemistry on fuel cladding behavior of
LWR in Japan," in: IAEA Specialist's Meeting, Leningrad, June 6-10, 1983 (1983), pp.
17-34.
2. C. Criss and J. Cobble, "'Tlie thermodynamic properties of high-temperature aqueous solu-
tions. IV. Entropies of the ions up to 200? and the correspondence principle,"
J. Am. Chem. Soc., 86, 5385-5390 (1964).
3. I. L. Khodakovskii, "Thermodynamics of water solutions of electrolytes at high tem-
peratures (entropy of ions in water solutions at high temperatures)," Geokhimiya, No.
1, 57-63 (1969).
4. C. Chen and K. Aral, "A computer program for constructing stability diagrams in aqueous
solutions at elevated temperatures,'' Corrosion NACE, 38, No. 4, 183-190 (1982).
5. P. Tremaine and S. Goldman, "Calculation of Gibbs free energies of aqueous electrolytes
to 350? from an electrostatic model for ionic hydration," J. Phys. Chem., 82, No. 21,
2317-2321 (1978).
6. U. Sen, "Study of electrolytic solution process using the scaled-particle theory. Part
3. Effects of thermal dilution on standard thermodynamic functions," J. Chem. Soc.,
Faraday Trans. I, 77, 2883-2899 (1981).
7. B. N. Ryzhenko, Thermodynamics of Equilibria in Hydrothermal Solutions [in Russian],
Nauka, Moscow (1981).
8. V. I. Zarembo and L. V. Puchkov, "Standard values of the Gibbs energy of formation
of ions and ionic associates in a water solution with high state parameters," Reviews
on Thermophysical Properties of Materials/TFTs, No. 2 (46) (1984)..
9. R. Mesmer and C. Baes, "Phosphoric acid dissociation equilibria in aqueous solution.to
300?C," J. Solut. Chem,, 3, No. 4, 307-322 (1974).
10. G. B. Naumov, B. N. Ryzhenko, and I. L. Khodakovskii, Handbook of Thermodynamic Quanti-
ties [in Russian], Atomizdat, Moscow (1971).
11. E. A. Melvin-Kh'yuz, Equilibrium and Kinetics of Reactions in Solutions [Russian trans-
lation], Khimiya, Moscow (1975).
12. H. Helgeson, D. Kirkham, and G. Flowers, "Theoretical prediction of the thermodynamic
behaviour of aqueous electrolytes at high pressures and temepratures. IV," Am. J.
Sci., 218, No. 10, 1249-1516 (1981).
13. V. S. Belyanin, "Study of thermodynamic properties of water iron compounds," Reviews
on Thermophysical Properties of Materials/TFTs, No. 4 (36), 109-166 (1982).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
PROBLEMS OF CHEMICAL-ANALYTICAL MONITORING IN NUCLEAR POWER
L. N. Moskvin UDC 621.039
Further increase of the reliability and efficiency of the operation of nuclear power
plants depends on the comprehensive solution of many problems, of which the problem of
optimization and increase of the informativeness of chemicotechnological monitoring is gain-
ing in importance. There is a clear rift between the level of scientific ideas and tech-
nical solutions on which nuclear power plants are based and the existing approach to the
organization of chemical monitoring. Everyone understands the necessity of chemical moni-
toring, but many people do not regard it as a necessary element for ensuring normal opera-
tion of a nuclear power plant. The lack of standard requirements and a standard methodical
base markedly lowers the reliability of the results of analyses, and behind this lies possible
breakdowns of the water-chemical conditions.
How can this situation be explained? The consequences of deviations of water-chemical
conditions from regulation standards are by no means manifested immediately. Apparent well-
being at any given time creates the impression that the requirements of the chemical laws
can be ignored with impunity. But there is also another reason for the disdainful attitude
toward chemical monitoring. Chemical-technological monitoring at a nuclear power plant
reduces to the determination of ti20 water-quality indicators for the basic and auxiliary
systems. Considering the number of existing points at which samples are extracted in
each block of a nuclear power plant and the frequency .with which the analyses are per-
formed, it is not difficult to imagine the impressive number of the overall volume
of data obtained. Thus the total number of analyses per month for one block of a nuclear
power plant with RBMK-1000 approaches 15,000 [1]. In addition, as a rule, these are single
measurements, whose error it is virtually impossible to estimate exactly. Moreover, regard-
less of how conscientious the chemists-analysts at the technological laboratories are, in
the monotony of repeating values it is difficult to escape big blunders in the case of unfore-
seen deviations of the parameters in the water-chemical conditions. Under these conditions
there is no hope of obtaining reliable information for each of the measurements, making sense
of the observational results, and drawing correct conclusions. We arrive at a paradox.
By increasing the number of parameters monitored and the number of points at which samples
are extracted we strive to increase the information content of chemical monitoring, but
we actually achieve the opposite result.
At the present time, when nuclear power has transformed from a unique source of energy
to one of the most important elements of the power production in the country, it is essen-
tial to reexamine the concepts forming the basis for the implementation of chemical-analyti-
cal monitoring at nuclear power plants. At the first nuclear power plants the research
and technological functions of analytical monitoring were balanced, and preference was often
given to obtaining research information. For serially produced nuclear power plants re-
search programs are a rare episode. It is evident that the main reason for this situation
is hidden in the existing standards and requirements imposed on the chemical-monitoring
system at nuclear power plants. At the first stages of development of nuclear power the
striving toward performing as much analysis as possible and over the entire technolgoiial
loop was justified, whereas at the present time, as experience in operating nuclear power
plants in this country and abroad shows, the time has come to search for new approaches
to the problem of analytical monitoring of the quality of water heat carriers.
The successful solution of the problem of analytical monitoring is often linked primarily
with the instrumentation. But the number of methodical and instrumentation developments
continues to increase, and the chemical-technological monitoring remains as before one of
the laborious and unreliable elements in the overall chain of operational monitoring of
nuclear power plants. Fundamental restructuring of the overall scheme of such monitoring
is possible only based on automatic or, at least, automated means of chemical analysis.
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 398-401, December, 1985. Orig-
inal article submitted November 16, 1984.
0038-531X/85/5906-0962$09.50 ? 1986 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
But this is still a secondary question, one or the constituent parts or a program wnicn must
be implemented in order to achieve a qualitatively new level in the organization of an on-line,
informative, and effective chemical-technological monitoring program.
At the present time there is no clear understanding of the problems of chemical-techno-
logical monitoring at nuclear power plants; these problems are formulated in terms of the
evaluation of the corrosion state of the equipment and the presence of deposits in different
sections of loops, primarily, in the active zone [2]. Is this true? .The occurrence of
corrosion processes is a regular consequence of the maintenance of water conditions. The
establishment of strict relationships between the quality of the heat carrier and the rate
of conversion is a scientific research problem. As in any chemical-technological process,
in the chemical technology of a nuclear power plant the problem of monitoring reduces pri-
marily to obtaining reliable analytical information about the regulated parameters. In
addition, for nuclear power plants the consequences of optimal flow of chemical processes
are not only the minimum rate of corrosion of structural materials, but also problems of
safety, linked, for example, with the state of the systems for afterburning of the fulmi-
nating mixture for boiling water reactors.
These problems predetermine the approach to the solution of the main problem. Either
the operating personnel are reponsible for solving the extremely complicated problem of
evaluating and forecasting the corrosion environment, which at the present time can by no
means always be solved by specialists in the area of corrosion of structural materials,
or they are required only to obtain reliable information on the parameters of those pro-
cesses to which they can react in real time. Here it is very important to understand the
particular parameters of the process which the operating personnel can affect and which
parameters are a consequence of technological failures or intraloop processes and cannot
be controlled on-line. For example, some indicators of the quality of the water heat
carrier (specific electrical conductivity, pH, concentration of sodium and chlorine ions,
corrective additives for correcting the conditions) can be regulated by operating personnel
when their values move outside the regulated zone, or technical meausres which prevent the
values of the parameters from changing substantially in the heat carrier can be adopted.
At the same time, some monitored indicators correspond to intraloop physicochemical processes
which the operating personnel can affect only indirectly through the change in the parameters
listed above. Thus two types of indicattrs can be distinguished: regulatable and informa-
tive. This separation enables, by understanding the cause-effect links, simplifying and
in some cases lowering the volume of monitoring based on the number of controllable indicators
and points of sample extraction as well as on the frequency of on-line monitoring. For
example, a significant fraction of the total labor involved in chemical analysis goes into
determining the concentration of iron and copper. And it is precisely these indicators,'
as a rule, which are the least reliable, since the content of these elements is often at
the limits of detection by suitable methods; in addition, the probability of errors owing
to random impurities is maximum.
Today the formulation of the problem itself could be surprising: is it necessary to
monitor iron and copper within the framework of technological monitoring? But let us con- .
Sider what the purpose of these indicators is. Observing the high content of iron in water
in the process of prestartup flushing, during the startup period or during the operation
of the plant, the operating personnel wait until it drops, since they observe the devia-
tions of the water conditions which led to this jump a long time ago on the basis of other
parameters and took appropriate measures. In none of the possible situations does an indi-
cator such as the concentration of iron require on-line interference in the technological
process, i.e., this is a typical informative indicator.' And since it is informative, it
is possible and necessary to decrease the volume of monitoring of the iron concentration
right down to complete elimination of this parameter as our depth of understanding of intra-
loop processes increases. The same can be said about copper.
A substantial reduction in the number of analyses performed can apparently be achieved
by taking into account the internal interrelationship of the indicators of the quality of
the water heat carrier and not only of it, but also of the cooling water. Why, for example,
should the "hardness" of the vapor condensate be determined? The presence of leaks can
be judged from' indicators such as the concentration of sodium and chlorine ions. If, on
the other hand, it is necessary to know the exact content of "hardness salts," then it is
simpler to determine the ratio of the sodium and calcium concentrations in the cooling water,
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
whll.ll, 6J a LlL1G, LG111CL 111J 1~V11J LOllL 1V1 LGJCLVVII l.VV1G111., Gull l.V VCLCL1I11I1e l.Ile ilGrllIle55
by a computational method using the correlation coefficient found. At the present time
it is difficult to give practical recommendations for taking into account the existing corre-
lations in the nature of the changes of different parameters of the water heat carrier, since
there are few studies of this question because of the inadequate reliability of the primary
information, obtained, as a rule, with the use of simple-extractive methods of analysis,
which are distinguished by low metrological characteristics.
Very often, especially when laboratory methods of analysis are discussed, the determina-
tion of the content of one or another component of the water solution is unreliable because
the essence of the methods used are not understood or because of the fact that any method
of determination, even the best method, is inherently uncertain and ignored. Several examples
can be presented. At many nuclear power plants the nephelometric method with a lower limit
of determination of ti25 ug/1 is used to determine the concentration of chloride ions. De-
tailed studies [3] have shown, however, that in order to achieve the indicated limit. a sample
volume of riot less than 400 ml is required, but even in this case the reliability of deter-
mination does not exceed 50~. Another example is associated with the determination of the
specific electrical conductivity and the pH. Quite often these indicaotrs are determined
by sample-extraction methods, forgetting that in this case contact with the atmosphere changes
these indicators in an uncontrollable fashion because carbon dioxide dissolves in the samples
extracted. The use of automatic instruments without proper maintenance also does not exclude
the possibility of the appearance of large errors. These errors are most often caused by
the inability to perform correctly the primary calibration of the apparatus. For example,
in the method for calibrating the "pNa-meter" proposed in the technical documentation for
this device, it appears that the residual content of sodium atoms in the reference water
is taken into account, but~it is not clear how this residual content is determined. Finally,
a quite prevalent rough error is the use of only one determination as the result of the
analysis. It is evident that it is either necessary to know the reproducibility and the
accuracy of the analysis and give the results with the corresponding error or to perform
a series of parallel measurements and to determine the average value. In this sense reduc-
tion of the volume of monitoring will enable meeting more strictly the metrological require-
ments in performance of laboratory analyses. The most important result of the reduction
of the volume of chemical monitoring as a whole is the possibility of complex automation
and, as a consequence, raising the relibility and validity of the results obtained.
The transition from manual sample-extracting methods of monitoring to automatic moni-
toring means not only continuous acquisition of data, but also the possibility of utilizing
the data for forecasting and determining the reasons for deviations from fixed water condi-
tions and for well-founded on-line interference in the technological process. Experience
in operating automatic chlorine meters at nuclear power plants shows that the change in
the concentration of chloride ions can be recorded earlier than the change in the electri-
cal conductivity. This is linked to the fact that the specific conductivity is an integral
indicator, to which the most highly mobile hydroxyl ions and activated protons make the
main contribution. From here it follows that when selecting the means for monitoring it
is primarily necessary to develop sensors which react selectively to definite impurities.
There now exists an instrumental-methodical foundation for continuous monitoring of the
most important parameters of the coolant in the flow under correction-free water conditions.
The changeover to automated chemical monitoring systems is inseparably linked with
the introduction of automated systems for processing of the results of analysis. In addi-
tion, such systems must not only fix the entire set of data and compare the results accord-
ing to the times and points at which samples are extracted, but it must also have the
capability of forecasting the possible flow of the technological process, as well as pro-
viding information on the reasons for the breakdown of the water conditions.
This closes the logical chain. The improvement of the chemical technology of nuclear
power plants opens up the possibility of reducing the volume of chemical monitoring. The
minimum number of monitored parameters is an insurance for reliability of the results and
opens up a real possibility for full automation of chemical monitoring. Automated means
of chemical monitoring based on flow through sensors will maximize the effectiveness of
analytical information on the technological process. From here follows the conclusion that
the most important problems of chemical-analytical technological monitoring at nuclear power
plants now lie at the boundary with technological problems and cannot be solved only by
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
chemists-analysts. tt is airricutt to expect signin cant progress in tnis question, it
only the methods and means of chemical monitoring are improved.
LITERATURE CITED
1. E. P. Kazakova, V. A. Mamet, and V. F. Tyapkov, Nuclear Power Plants [in Russian],
No. 2 (1979), pp. 180-183. "
2. 0. I. Martynova, L. M. Zhivilova, and N. P. Subbotina, Chemical Monitoring of the Water
Conditions in Nuclear Power Plants [in Russian], Atomizdat, Moscow (1980).
3. Yu. M. Kostrikin, Teploenergetika, No. 1, 52-54 (1976).
REMOVAL OF CORROSION PRODUCTS FROM THE STEEL SURFACE IN THE AQUEOUS
COOLANT OF NUCLEAR POWER PLANTS
V. G. Kritskii, A. S. Korolev, UDC 621.039.553.36
I. G. Berezina, and M. V. Sof'in
Corrosion of the materials of the low-pressure preheater tubes and casings, piping,
and other elements of the condensate supply channel of NPP does not usually affect their
operational reliability and life. However., during operation, a part of the products passes
into water, and is subsequently transported to the reactor and enters the primary circuit.
In boiling water reactors, the corrosion products settle mainly on the surface of the fuel
elements. They can cause damage to the fuel elements and, after activation, they are~dis-
tributed along the circuit surfaces and significantly increase the total level of the coolant
activity and the radiation dose from the system. A knowledge of the conditions under which
there is an increased removal of the corrosion products helps one to considerably decrease
entry of the corrosion products into the primary circuit of the reactor.
This paper deals with the investigation on the effect of different factors on the trans-
fer of the corrosion products of steels into the aqueous medium. The kinetics of corrosion
and removal of the corrosion products was studied on the specimens tested under the condi-
tions of the condensate supply channel of a NPP having RBMK reactors and under static con-
ditions. The indicator-specimens were placed in the deaerator tanks, deaerator column (in
the decontaminated condensate-stream), and in the mechanical filtration unit of supply water.
The specimens were withdrawn after testing for 3200, 5000, and 9000 h. The quality of water
of the condensate supply channel met the specification OST 95-743-79. In one experiment,
we monitored the oxygen content in the decontaminated condensate, and under static condi-
tions - the content of the oxidizing agent H2O2 in water. The treatment of the specimens
before and after testing, and the calculations of the corrosion rate and the rate of removal
of the corrosion products were carried out according to the procedure described elsewhere
[1] (Table 1).
The magnitudes of corrosion and removal of the corrosion products in the supply water
are described by the so-called "sigmoidal" curves (Fig. 1). The initial incubation period
is particularly noticeable in the case of the Kh18N10T steel. In the case where the quantity
of the corrosion products retained in the specimens exceeded the quantity calculated on
the basis of their weight loss after removing the films, we considered that the precipitation
processes of the corrosion products from water took place.
In the conventional method of evaluating the magnitude of removal of the corrosion
products (1, 2], the degree of transfer of the corrosion products into water has been estab-
lished using the ratio of the specific weight of the product entering water and the specific
weight of all the corrosion products of the steel formed under the given conditions as the
criterion. It was found that the removal of the corrosion products varies smoothly (con-
tinuously) with time and depends on the chromium content in the alloy. Such a trend of
the curves is less informative. In the computed models, the tabulated values of the per-
centage removal for each grade of steel (under different conditions) are simply assigned.
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 401-403, December, 1985. Orig-
inal article submitted November 16, 1984.
0038-531X/85/5906-0965$09.50 p L986 Plenum Publishing Corporation 965
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
?O10
C
y 70
O
U0 i 2 3 k 567 B 910~,103h ~
p ~ 70
a~
A coo _
Fig. 1 Fig. 2
Fig. 1. Corrosion (a), and deposition and removal of the corrosion products
(b) in the supply water: ~) steel 20; C)) steel Kh18N10T.
Fig. 2. Relationship between the magnitudes of corrosion and the removal
of corrosion products observed in the tests in two water-chemical (hydro-
chemical) regimes: 1, 3) A = 1; 0.5; 2) oxygen-free water; 4) water with
oxygen (40-300 ug/kg); ?) supply water (test duration: 4750 h, June
1983); ~) deaerator No. 51 (tank) (3262 h, June 1982); x) deaerator No.
61 (tank) (3240 h, June 1982); ^) deaerator No. 21 (tank) (8950 h, December
1981); ?) multiple forced-circulation loop (8030 h, December 1981); O)
deaerator No. 61 (column) (3240 h, June 1982); ?) deaerator No. 51 (column)
(3262 h, June 1982).
TABLE 1. Chemical Composition of the
Experimental Steels
Melt
number
8
9
16
17
68
70
305
307
3
4
Kh18N10T
73
AS -9
205
'L2i{
'Phere is a complete. change in
o,os
n,o8
0,08
0,08
0,10
0,14
0,11
0,08
(1, 03
0,03
0,37
0,29
0,17
0,21
0,3
0,3
0;46
0,38
0,37
0,41
i
0,40
~0, 37
,0, 26
0,26
0,6
0,6.
0,52
(1,52
0,36
0,44
21,7
13,0
4,8
9,0
0,03
1,50
o,oss
0,19
111,0
24,2
0,25
1,05
0,24
O,G7
O,Gfi
2,1
0,1
0,38
0,010
0,011
0,021
0,020
Standard composition
0,3 X0,3 10,6 1 2,6 I - I - 10,01010,021
Standard composition
0,03I0,29I0,23I11,6 I - I0,05I0.017I0,029
0,'L60,32 0,77 0.98
the pattern when the data of the
plotted in different coordinates: the absolute values
and the removal (deposition) of the corrosion products
metal content in the corrosion products on the y axis.
of corrosion
recalculated
For a series
corrosion tests are
on the x axis, g/m2,
with respect to the
of specimens, the re-
lationship between corrosion and removal (which, in turn, are nonlinear
and the degree of alloying) can be described by the following linear equation (Fig. 2)
where B is the removal within the test duration, g/m2; K is the corrosion within this period,
g/mZ; Ko is a constant that may be interpreted as the weight of the metal contained in the
minimum protective layer under the conditions of testing, g/m2; and A is a constant that
depends on the type of water-chemical (hydrochemical) regime.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
equation ~1J snows tnat the =itm thickness ~S can oe iouriu our. arum sue eyua~icns A~ _
K - B; 0~ = K - AK + AKo. If AKo = ~~o-_the initial thickness of the protective film - then
~~ - ~~o = K(1 - A). This experimentally established fact was verified on the published
data for steels in all the water-chemical regimes that are of interest in the nuclear power
engineering. In this case, the following values were established for the constant A: A
1 (for the primary circuit of the PWR reactors); A ~ 0.5 (for the primary circuits of the
BWR and RBMK reactors); A ~ 0.66 (for deaerated supply water having pH ~ 7); and A'< 0.5,
usually ti0.25 [for the decontaminated condensate with pH ^' 7 and with dosing of oxidizing
agents (air, 0?, or HZ02)].
The steels having different chromium contents usually have the same value of A in a
given water-chemical regime; on the other hand, the change in the magnitude of removal with
time is a phenomenon apparently related'to the duration required for the formation of the
initial protective oxide film (see Fig. 2). In the general case, the topochemical reaction
of the formation (growth) of the protective surface layer takes place at first, and is
followed by a transition to the diffusion-controlled (through the already formed layer)
kinetics.
The growth rate of the oxide layer is usually described by the following differential
equation [3]
where ~r is the layer thickness; and Kp is the parabolic-growth constant. If we assume
that the removal of the corrosion products is accomplished independent of ~~ and time (ero-
sive wash-off ), .the resulting rate can be written as
where C is the constant of the erosion process. In case the removal of the corrosion prod-
ucts is due to the redistribution of the incoming ions from the metal between the oxide
film and the solution, the growth rate of the film is given by
where A is the constant of removal as obtained from Eq. (1). Integration of the Eqs. (2)-(4)
gives
i= K 0~2~-const;
P
ti= Kp [~~3-4tg1 + 3 1 K2 1 ~O~s-oil-f ...;
ti= K ((1-A)~O~-I-Q~ol~,
P
respectively, where ~~o is the initial thickness of the film at T = 0. At C = 0 and A =
0, Eqs. (6) and (7) transform into Eq. (5).
Using the coordinates 4g = f (4~) and computer analysis of the results, the values
of all the constants entering Eqs. (5)-(7) are generally obtained. However, in this case,
one requires the value of the coefficient A obtained when determining the magntiude of the
removal (transfer) of the corrosion products into water. Therefore, it is recommended that
one must not only record the corrosion (metal) losses, but also study the film. This is
particularly important in view of the fact that in neutral media (according to the results
of numerous experimental investigations) there is a correlation between the quantity of
the material settled in the external 'loose' (porous) layer and that settled in the dense
film. The complex approach adopted to determine the redistribution of the metal between
the oxide layer and the coolant during the corrosion process of the metal shows that this
process is not accidental and that it depends on the properties of the coolant.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
LITERATURE CITED
1. V. M. Nikitin, A. M. Gvozd', and T. Ya. Karpova, "Regularities in the transition of
the corrosion products of steels into the aqueous media," Teploenergetika, No. 8, 44-48
(1981).
2. I. K. Morozova, A. I. Gromova, V. V. Gerasimov, et a1., Removal and Deposition of the
Corrosion Products of the Reactor Materials [in Russian], Atomizdat (19?5), p. 280.
3. K. Hauffe, Reactions in Solids and on Their Surfaces [Russian translation], Part II,
Izd. Inostr. Lit., Moscow (1963).
0. T. Konovalova, M. I. Ryabov, UDC 621.039.548.5
L. N. Karakhan'yan, and T. I. Kosheleva
In order to establish the characteristics of the water (aqueous) regime and the means
of maintaining it, it is necessary to determine the concentration of the corrosion products
in the multiple forced-circulation loop (MFCL) of a boiling water channel reactor. The
quantity of corrosion products of iron (c.p.i) entering the coolant from the i-th segment
of the loop (circuit) per unit time can be expressed as
Bt = K~Pes~ti o.s (1)
where KiT-0.5 is the corrosion rate, g/(m2?h); pi, fraction of the corrosion products enter-
ing the coolant; Si, surface area of the segment of the loop, mZ; and T, time, h. The total
quantity of the corrosion products entering MFCL is given by
n n
8=0.5 ~' BZ=0.5(~ K:P~~S~)i-o.s. (2)
t i
The coefficient 0.5 takes into account the degree of decontamination from the corrosion
products during the condensate purification treatment.
During decontamination of the loop, G1 gram corrosion products are removed per hour:
G1 = 0.5PC, where 0.5 represents the degree of decontamination; P is the coolant consumption
for decontaminating the loop, kg/h; and C is the concentration of the corrosion products,
g/kg.
Steam carries away G2 = K1NC corrosion products, where K1 is the coefficient of dis-
tribution of the corrosion products between water and steam; and N is the productivity of
steam, kg/h.
On the surface of the heat .liberating elements one observes deposition (settling) of
the corrosion products amounting to G3 = O.StK2C, where 0.5 is the coefficient of nonuni-
formity of deposition along the length of the fuel. element; St is the total surface area of
the fuel elements, m2; and KZ is the coefficient of deposition of the corrosion products
in the area of boundary (wall) layer boiling. The balance equation of the corrosion prod-
ucts assumes the following form:
n
0.5 (~' KiP~S~) .~-o.s = (0.5P -~- K,N -}- 0.5 SLKZ) C.
1
In the case of a boiling water channel reactor, the equation has the following form
]gC= -3.6-0.51gi.
We experimentally obtained the following time dependence of the concentration of the
corrosion products
1gC= -(3-4)-0.51gti. (5)
Translated from Atomnaya ~nergiya, Vol. 59, No. 6, pp. 403-405, December, 1985. Orig-
inal article submitted November 16, 1984.
0038-531X/85/5906-0968$09.50 p 1986 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
1000~`~~ d?
400 ? .?~ ~~.
200 ; ???~ ~~.
~uu
70
40
U ~
~~
0
. ? . ~ ~
????
4 7 10 20 40 70 100200 400 1000 4000 10000
Campaign period, h
~ 12s
~ 100
7s
x
icy 50
25
.y
O
N ~
A
5000 10000 ' 15000
Campaign period, h
Fig. 1 Fig? Z
Fig. 1. Calculated (---) and experimental (?) values of the iron
concentration in the MFCL of~a boiling water reactor [2].
Fig. 2. Calculated (1) and experimental (2) values of the deposits
on the fuel elements of the boiling water reactor [3].
Along with the aforementioned method of calculating the concentration of c.p.i in reactor
water, there was a parallel development of another method of calculating the mass transfer
and the maximum deposition of c.p.i based on the theory of formation of deposits (precipi-
tates) [1] according to which in the general case, the deposition mechanisms of c.p.i are
described by different mathematical relationships in the regions of single phase stream
with convective heat exchange, boundary layer and developed bubbly boiling, and in the zone
of cyclic flow regime of the steam-water mixture. The c.p.i can exist in reactor water
in two forms: dissolved and undissolved. Both forms of c.p.i. precipitate on the fuel
elements, but the mechanisms of their deposition are different. The quantitative relation-
ship between the concentrations of the dissolved and the undissolved c.p.i depends on the
temperature and the water (aqueous) regime.
According to the stated method, the time dependent concentration of c.p.i is found
out based on the balance between the entrance-and the removal of c.p.i. into the reactor
water:
C- 1C ~, B-St %p-NK1Cp 1111 + xP-{ Nag ~ ~
t p St.K~P-I-SK*P Sc Fft*P-I-SK*P
(6)
where Cp is the concentration of the dissolved c.p.i, g/kg; S, area of the unheated surfaces
of MFCL, m2; gp, average deposition rate of the dissolved c.p.i on the fuel elements accord-
ing to the crystallization mechanism, g/(m2?h); Kt and Ks?, average deposition coefficients
of the particles of c.p.i on the fuel elements and on the unheated surfaces, m/h; p, coolant
density, kg/m3; Ek, coefficient of dropwise removal; and x, coefficient of effectiveness
(efficiency) of the decontamination system.
Using the known concentration of c.p.i in MFCL, one can find out the values of de-
posits on the fuel elements and on the unheated surfaces, and extraction of iron by the
decontamination system. For example, the quantity of deposits on a local fuel element
segment G (g/m2) is given by:
C;-g i-}- ~ K1~C-Cp)Pdi=g i-}- ~*p {0.5Bo ~t-t[~P~xl'-I-Neg-~-1VK~)-I-STSP1}
P o P Stlfr*P-I-3K*P-I-xP-I-n~Ek ,
(7)
where the first term takes the deposition of the dissolved c.p.i into account and the second
term (integral) accounts for the deposition of the particles of c.p.i; gp is the local growth
rate of the deposits of dissolved c.p.i on the fuel elements, g/(m2?h); Kt* is the local co-
efficient of deposition of the particles of c.p.i on the fuel elements, m/h; and Bo =
EKipiSi. All the parameters entering Eqs. (6) and (7) can be determined quantitatively using
the previously published relationships [1].
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
rigure 1 snows a comparison or the calculates ana the experimental values oT the iron
concentration in the MFCL of a boiling water reactor.
During the intial period of campaign (during which one observed the maximum growth
of deposits), the calculated values of the iron concentration lie close to the upper limit
of the experimental values. After operating the reactor for 5000 h, the level of the experi-
mental values`of the iron concentration sometimes exceeds the level of the calculated values.
Apparently, this fact is related to the periodic erosion (wash off) of the deposits because
of the transient periods and other circumstances.
Figure 2 shows a comparison of the calculated and the experimental values of the deposits
on fuel elements of the boiling water reactor.. After operating for 2 years and acid washing,
the thickness of the deposits in the zone of the boundary layer boiling reaches a level
of 100 }un;.the calculated curve also indicates this level. For the purpose of calculations,
the density of the deposits was taken as 2 g/cm3.
In the experiments, we recorded the local deposition of c.p.i under the spacer grids
(SG) in the fuel elements assemblies of the boiling water reactors which exceeds-the deposi-
tion on the adjacent regions of the fuel elements by 1.3-5 times. In this case, as one
approaches the exit end of the assembly (i.e., with increasing flow rate of the steam-water
mixture), there is a relative growth (increase) of the deposits under the grids.
The effect of SG may be explained in the following way. They hydraulically act on
the coolant flow such that it partially deviates from the axis and streams directed towards
the fuel elements develop in it. During this process, all the particles of c.p.i moving
in such a stream reach the fuel elements surface overcoming the boundary layers of water
due to inertia. The growth rate of the deposits under the action of such a stream gSG~
g/(m2?h), is given by
gsc = 36000w Wpb, ( 8 )
where Cw is the concentration of the particles of c.p.i in water, g/kg; W, local flow rate,
m/sec; ~ = exp(-E/RT), probability of adherence of the particles of c.p.i to the wall; and
E, activation energy of the surface dehydration process of the particles of c.p.i, kJ/kg.
According to Gerasimov [1], ~ z 10-4. The speed (flow rate) of the coolant washing
the assembly increases along its height from 2 up to 20 m/sec. The local coefficients of
the deposition process of the particles of c.p.i (KSG, m/h) on the fuel elements under the
influence of SG are equal to KSG 3600W~ = 3600 (2-20)10-4 .l: 0.5-10.
The local coefficients of deposition on the fuel elements under the influence of the
other processes examined earlier [1] are equal to Kxt = 0.2-15 m/h. As one approaches the
upper end of the assembly, Kt values decrease because of the reduced thermal flux.
Thus, SG can make the coolant stream deviate towards the wall, and, thereby, cause
additional growth of the deposits that is comparable to the effect of other factors.- This
relative contribution of the effect of SG on the deposition process must increase with in-
creasing flow rate; this agrees with the experimental data. In order to avoid increased
deposition under SG, it is necessary to decrease the total concentration of iron in reactor
water.
LITERATURE CITED
1.
V.
V. Garasimov, Corrosion of Reactor Material [in Russian], Atomizdat, Moscow (1980),
pp.
163-181.
2.
A.
the
I. Gromova and V. P. Sentyurev, "USSR-UK Seminar on the water-chemical regimes and
structural materials of boiling water channel reactors," At. Energ., 45, No. 1,
77 (1978).
3.
I. A. Varovin, A. P. Eperin, M. P. Umanets, and V. G. Shcherbina, "A decade-long ex-
perience on the operation of the Leningrad NPP," ibid., 55, No. 6, 349 (1983).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
FORMATION OF DEPOSITS ON THE SURFACE OF THE FUEL ELEMENTS
OF RBMK-1000
I. A. Varovin, S. A. Nikiforov,
A. P. Eperin, Yu. N. Aniskin,
V..G. Kritskii, and Yu. A. Khitrov
UDC 621.039.553.36
At the present time, it is considered that the phenomenon of particle binding on the
surface of the fuel elements directly leads to the formation of the structures of constant
density. Here, the particle binding stage is visualized as the crystallization of the dis-
solved substance (form) and dehydration of the dispersed particles (d.p.) on the surface
(1]. The following observations (that are useful for ensuring reliable operation of the
fuel elements) reflect this concept: safe operation of the fuel elements is determined
only by the effect of the total thickness of the deposit (precipitate) layer; the coef-
ficient of thermal conductivity of the deposits remains constant.
There is yet another possible theoretical interpretation for the phenomenon of d.p.
deposition. Their binding is a consequence of the initial formation and continuous densifi-
cation of the low-density structures.- In this case, the coefficient of thermal conductivity
of the deposits becomes a variable that depends on the rate of formation of the layer of
the more porous (loose) structure and its transformation into a more dense. structure and,
also, on the d.p. concentration in water and the operational duration of the power system.
Under these conditions, safe operation of the fuel elements may not be determined by the
total thickness of the deposit layer, but it may depend on the porous (loose) portion of
this layer that has the maximum thermal resistance.
The published values of the average coefficients of thermal conductivity of the deposit
layer vary over a wide range: 8.6-0.015 kcal/(m?h?deg) (2, 3J. It is difficult to explain
such a large discrepancy of the experimental data if the formation of the constant-density
deposits occurs directly. At the same time, it is natural if we note that the average co-
efficient of thermal conductivity of the deposits is a function of the relationship between
the layer thickness and the duration of existence of the deposits. In view of this, the
concept of variable density of the deposit layer is fairly admissible. A special study
is required for answering the question: which of the two aforementioned hypotheses reflects
the physical essence of the processes occuring under the actual conditions?
FOR THE FORMATION OF DEPOSITS
Separation (precipitation) of the deposits from the solutions containing d.p. is often
a result of the formation of the. periodic colloidal structures (p.c.s.) which is an inter-
mediate stage of the formation of dense deposits during the evolution process of the elec-
trophoretic deposits of d.p. At the present time, it has been shown [4] that the polarizing
interaction of the particles and the possibility of their subsequent aggregation must be
taken into acocunt when studying the evolution processes of the electrophoretic deposits.
The dominating effect"of the polarizing interaction during the formation of p.c.s. offers
a possibility for the separation of dense coatings owing to comparatively easy sliding:
of the particles and their aggregates (chains) relative to each other.
The hydroxides entering the composition of the dispersed products have a significant
polarizing dipole moment [4]. Electric fields form in the boundary liquid layer also. These
fields form due to the corrosion processes and thermal flux. Heat transfer through the
boundary layer requires temperature gradients inducing thermal diffusional processes as
well as diffusion of the dissolved form of the barely soluble compounds whose solubility
decreases with temperature variations.
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 405-409, December, 1985. Orig-
inal article submitted November 16, 1984.
0038-531X/85/5906-0971$09,50 ? 1986 Plenum Publishing Corporation -971
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
The formation of the electrical fields is related to the differences in the ionic mo-
bility during the diffusion and thermal diffusion processes and, to the tendency of the
solution to remain electrically neutral. This internal electric field accelex'ates the slower
ions and decelerates the faster ones. In a binary electrolyte solution it ensures identical
ion transfer rate during the diffusion and thermal diffusion process. The electric poten-
tial difference appearing in the liquid is one of the components of the diffusion and thermal
diffusion potentials that can be measured using electrochemical methods [5, 6].
The existence of the electric field in water and the dipole moments in d.p. is a suf-
ficient condition for the formation of the layer of the variable-density deposits as a result
of the formation of p.c.s. They form in the liquid boundary layers of the power systems.
Now, the question is: how significant is the effect of this field on the structure evolu-
tion of the deposits? The answer to this question requires determination of the boundary-
layer electric field intensity, and also, the intrinsic and polarizing dipole moments of
the particles. At the present time, direct measurement of the component of the diffusion
and thermal diffusion potentials (having the physical meaning of an internal electric field
in the electrolyte solution) remains an unsolved experimental problem [6]. In certain cases,
it may be evaluated theoretically; for example, it is possible to calculate the electric
fields developing in dilute solutions under the influence of the diffusional processes.
In the isothermal liquid layer, calculating the electric field intensity of the dilute solu-
tions is complicated because of the absence of the data on the coefficients of thermal diffu-
sion of ions. In view of this, it is difficult to give a theoretical evidence for the pos-
sibility of deposit formation of a result of the development of p.c.s.
At the same time, there are experimental data which can be interpreted as a confirmation
of the suggestion that in certain cases, the process of deposit formation on the heat
transferring surfaces occurs precisely in this way. In fact, when the heat flux decreases
significantly, reversal of the corrosion products is often observed (2, 7]. A change in
the heat flux causes a change in the electric field intensity in the heat-transferring liquid
layer and a corresponding change in the polarizing interaction forces. Bond weakening within
p.c.s. results in the hydrodynamic separation of certain d.p.
However, these experimental data can be interpreted differently: the reversal of the
corrosion products under reduced thermal flux is a consequence of the increased solubility
of magnetite with decreasing temperature [2] and, therefore, the observed experimental fact
must not be considered as a direct proof for the formation of the deposits through p.c.s.
EXPERIMENTAL CONFIRMATION OF THE DEPOSIT FORMATION
THROUGH THE STAGE OF THE PERIODIC COLLOIDAL
STRUCTURE EVOLUTION
A theoretical examination of the process of deposit formation on the heat transferring
surfaces shows that at the present time, there are, in principle, two approaches available
for understanding this aspect.
1. The deposit formation process consists of two stages: supply (admission) and bind-
ing (pinning). The relationship between them is realized only through the d.p. concentra-
tion. The binding stage is related to their dehydration.
2. During the formation of deposits, there is an intermediate stage of p.c.s. evolu-
tion between the stages of dehydration and supply. The relationship between the supply
and the formation of p.c.s. may be accomplished through the d.p. concentration as well as
through the driving force of these processes.
The driving force of the process of d.p. transfer through the liquid boundary layer
and the formation of p.c.s. is the electric field developing in this layer in the presence
of temperature gradients. Based on an analysis of the theoretical and experimental de-
pendences of the rate of deposit formation, one must not draw unequivocal conclusions regard-
ing the route of deposit formation: directly through the dehydration stage or through
a prior stage of p.c.s. formation. An answer to this question was obtained from an analysis
of the temperature variation in the fuel element jackets in the region of the surface boil-
ing zone of the thermometric fuel assemblies (FA).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
The deposits accumulating in the Tue1 elements Corm an aaaitional thermal resis~ance
between the fuel element jacket and the main flow of water, and therefore, under constant
thermal flux density, the formtion of deposits is invariably related to the temperature
increase in the fuel elements.
The structures developed by .the dehydrated particles and p.c.s. significantly differ
from the standpoint of thermal conductivity. Dehydrated d.p. form porous structures that
are close to the crystalline structures (magnetite type) and whose thermal conductivity
is virtually determined by thermal conductivity of magnetite taking its porosity into
account. The density of these structures (here again, taking the porosity into account)
is approximately a few grams per cubic centimeter. Their thermal conductivity depends on
the heat transfer zone in which they form. In the region of surface boiling, the thermal
conductivity of these deposits is considerably higher than that observed in the zone of
convective heat transfer and well-developed boiling [2]. According to these data, the ef-
fective thermal conductivity of the iron oxide deposits in the surface boiling zone amounts
to 8.6 kcal/(m?h?deg).
The p.c.s. have a low density because of considerable separation of d.p. from each
other. Their thermal conductivity may be taken as virtually equal to that of water 0.5
kcal/(m?h~deg). During heat transfer, the appearance of a p.c.s. layer is equivalent to
thickening of the lamellar layer. Thus, in the region of surface boiling, the deposits
forming directly through the stage of dehydration and through the intermediate stage of
the p.c.s. evolution can differently affect the temperature in the fuel element jackets.
The rate of formation of the deposits is governed by the rate of entrance (arrival) of
the corrosion products. The rate of out-flow of the corrosion products of the main materials
of the NPP circuits decreases with time according to the law: T-0's, where T is the opera-
tional time of the system. If the process of deposit formation occurs only through the de-
hydration stage, the jacket temperature at a point located in the region of the surface boil-
ing zone under a given channel capacity (power) must increase linearly according to Tp +
bT?'s, where T~ is the temperature at the point after the channel is set to the normal capa-
city; and bT?' is the temperature change during operation.
If the process of deposit formation takes place through the intermediate stage of p.c.s.
evolution, the thickness of the intermediate layer of p.c.s. depends on the d.p. supply and
dehydration rates. Depending on the ratio of these rates at the point located in the region
of possible existence of the surface boiling zone, a different nature of temperature varia-
tion must be observed in the fuel element jackets when the channel is working at a constant
capacity.
If the dehydration rate is substantially higher than. the d.p. supply rate, the tempera-
ture of the fuel element jackets increases as in the absence of the stage of p.c.s. formation,
i.e., as a function of the form Tp + bTO.s_ _
If the rate of p.c.s. formation is slightly higher than the dehydration rate, the tem-
perature of the fuel element jackets increases somewhat slower than that given by the func-
tion Tp + bT??s because of the reduced thickness of the p.c.s. layer as a result of the
decreased d.p. concentration in water with time.
If the rate of p.c.s. formation is much higher than the dehydration rate .and the thermal
conductivity of the hydrated deposits is much higher than that of the p.c.s. layer, then
the temeprature of the fuel element jackets will be maximum at the initial operational period
of the system during which the concentration is maximum, and it may decrease with time as
the system continues to operate, i.e., T = f(T, 8p,c.s.)?
Figure 1 shows that the maximum temperature of the internal jacket of the fuel- elements
decreases with time. This effect may be interpreted as a consequence of decreased capacity
(power) of the fuel assemblies because of fuel depletion and reduction in the thermal re-
sistance of the deposit layer. Using Fick's law, we can write
~TZ - a - 4abz~i _ Nasx~i (1)
~Tl - T Plg la'2 Nlsl~2 t
where q is the thermal flux density; b is the thickness of the deposit layer; .a is the co-
efficient of thermal conductivity; and N is the thermal capacity of the channel. When the
deposit layer forms at a rate that decreases in proportion to the square of -time and the
coefficient of thermal conductivity remains constant, Eq. (1) has the following form:
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
,_ ,
Tf0
Jd0
J20
Fig. 1. Maximum temperature of
the fuel element jacket of the
thermometric cassette in the
zone of surface boiling (cell
45-54, 1974 and 1977).
N
7800 2Q, 30 10 20 days
Fig. 2. Temperature increase in
the fuel element jacket of the
thermometric cassette when setting
the unit to the nominal capacity
within 3 days (August, September,
1978).
05
NQtiz.
wT- 05
NIti1?
The maximum temperature of the fuel element jackets was recorded in 1974 approximately
after 6 months operation of the apparatus. The fuel assembly (FA) was taken out in September,
1977, i.e., after operating for 39 months. During this period, the power of FA decreased
by less than Z times. For this case, from Eq. (2) we obtain aT > 1, i.e., the temperature
of the fuel element jackets (if the thermal resistance of the deposits remains constant)
must increase with increasing operational period of the system. Thus, the decrease in the
temperature of the fuel element jackets with the growth of deposits indicates that if the
deposits are not partially removed during operation, the thermal resistance of the layer
is not a constant and the main contribution to it comes not from the dense layer of the.
hydrated corrosion products, but from the layer of colloids forming p.c.s. whose thickness
decreases as a result of the reduction in the concentration of the corrosion products in
water of the power systems with time.
However, if the colloid layer forms a significant thermal resistance during the steady-
state operation of the system, then we expect that its formation can be observed even after
a prolonged shutdown. In this case, the temperature of the fuel element jacket at the point
located in the region of the surface boiling zone changes differently depending on the manner
in which the deposits form. If the process of deposit formation does not occur directly
through the dehydration stage, then, with increasing power of the system, the temperature
of the fuel element jacket is determined not only by the time required for setting the system
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 . Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9 ch
t0 1...~ ..~.t....,....~, .,.... ..~.,., .,~ .._...... .,_ _.,-......~_.,.. __ ._ ......_.._.._.-_~ r----- --~-- --- "---
the rate of entrance of d.p. is equal to their dehydration rate.
Figure 2 shows that the increase in the jacket temperature continued with some devia-
tions even after the system attained its nominal capacity on 11th August; the temperature
was stablized only on 20th August.- The temperature of the internal jacket of the fuel ele-
ment remained constant up to 10th September after which we observed its deviation from the
steady value. All the observed deviations are related to the changes in the channel power
and the coolant consumption through the channel during the reactor monitoring process.
Thus, the experimentally observed nature of the temperature variation in the fuel element
jackets is in support of the theoretical argument that the deposit layer formation occurs
through the stage of p.c.s. evolution. In view of the fact that iri the surface boiling
zone the main thermal resistance is offered by the p.c.s. layer and not by the dehydrated
layer, this conclusion is extremely important from the standpoint of ensuring reliable opera-
tion of the fuel elements having zirconium jackets in the boiling water reactors. The loca-
tion of the surface boiling zone in the reactor, the medium and the maximum values of the
thermal loads on the fuel elements, the permissible concentrations of the corrosion products
in reactor water, and the amplitude .and the frequency of vibrations (fluctuations) in the
surface boiling zone (when they are programmed in advance) must be chosen on the basis of
not only the total thickness of the deposits, but also. the thermal resistance of the layer
of colloids.
The significant effect of p.c.s. on the temperature of the fuel element jacket and
the fairly rapid response of this layer to the changes in the surface conditions permit
one to consider the temperature control of the jackets as an important part of the operative
control of the effectiveness of the measures taken for setting the water-chemical regime
and the start-up and the operational regimes of the system for improving the quality of
water in the multiple forced circulation loop (MFCL).
From the aforementioned facts it follows that:
during the formation of deposits on the fuel elements in the surface boiling zone,
the main thermal resistance can be offered by the layer of the hydrated colloidal-forms
with the p.c.s. formed during the first few days after putting the system into operation;
the permissible total thickness of the dense localized deposits that is regarded at
the present time as the unique condition for ensuring reliable operation of the fuel elements
is not an adequate criterion;
the temperature control of the fuel element jackets is an important part of the opera-
tive control of the effectiveness of the measures taken for setting the water-chemical regime
and the start-up and the operational regimes of the system for improving the quality of
water in MFCL.
LITERATURE CITED
1. V. V. Gerasimov, Corrosion of Reactor Materials [in Russian], Atomizdat, Moscow (1980).
2. I. K. Morozova, A. I. Gromova, and V. V. Gerasimov, Removal and Deposition of the Corro-
sion Products of Reactor Materials [in Russian], Atomizdat, Moscow (1975).
3. A. G. Rassokhin, L. P. Kabanov, S. A. Tevlin, and V. A. Tersin, "Thermal conductivity
of iron oxide deposits," Teploenergetika, No. 9, 12-15 (1973).
4. I. F. Efremov, Periodic Colloidal Structures [in Russian], Khimiya, Leningrad (1970).
5. J. Neumen, Electrochemical Systems [Russian translation], Mir, Moscow (1977), p. 464.
6. R. Haase, Thermodynamics of Irreversible Processes, Addison-Wesley (1968).
7. V. P. Brusakov, V. M. Sedov, K. D. Rogov, et al., "Regularities in the behavior of the
corrosion products in the NPP circuits," in: Interinstitute Reports of the Lensovet
LTI (1979).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
V. V. Gerasimov, A. I. Gromova, UDC 621.039.553.36:620..193.47.7
V. N. Baranov, and Yu. V. Makarenkov
During operation (1973-1984) of nuclear power plants (NPP) with RBMK-1000 reactors,
not a single failure was observed of controlled circulation loops (CCL) in 10 units, re-
sulting from the corrosive action of the coolant. So far at the Leningradskaya, Kurskaya,
Chernobyl'skaya, and Smolenskaya HPPs no underproduction of electric power has. happened
due to the damage caused by corrosion in the CCL or by large deposits of corrosion products
.leading to the failure of fuel .elements [1].
Reliable operation of the CCL equipment during a long period is a convincing proof
of the correct combination of the structural materials and reactor features, water chemistry
(operating conditions), and the corrosion protection means chosen. Therefore, the improve-
ment in the water regime is advisable when its advantages can be clearly substantiated for
the units and when the reliability of the loop equipment will not be impaired with its in-
troduction. As was shown in [2], the introduction of the oxygen regime in units of steam
power plants lead to the decrease in deposits on heat-transfer surfaces. No deposits causing
fuel-element damage were observed at NPPs with RBMK-1000 reactors [1], and the matter has
not been raised of a need to reduce them. .The advantage of the water chemistry is that
it provides, as we believe, the possibility to decrease the deposit formation along the cir-
cuit, affecting the radiation conditions in the CCL. We will therefore consider the reduc-
tion in corrosion products radioactivity in .the CCL, the Y radiation of which determines
the radiation conditions in the buildings and around the pieces of equipment in this loop
[3], for the improvement of the radiation conditions at the NPP leads in the long run to
a reduction in maintenance costs.
In accordance with design and experimental investigations of the radiation conditions
in the buildings and near the equipment of the main process loop and the radiation state
of this loop [3-6], for a substantial, e.g. 10-fold, decrease in the dosage rate near the
CCL equipment, the contents of the corrosion products of design materials in the coolant
(iron oxides, cobalt oxides, etc.) should also be decreased 10-fold. ~
In NPP. units with boiling water reactors, two variants of water chemistry are employed:
a neutral regime without correction and a correction regime with oxidant additives. As
oxidant, use is made of hydrogen peroxide (the Second Unit of the Beloyarskaya NPP; NPP(s)
in FRG) and gaseous oxygen (VK-50 [7, 12-14], NPPs in Sweden and Japan). The choice of
water chemistry for an NPP is dictated,as a rule, by the above factors and specific features
of plants. At the moment, in view of the circumstances mentioned, the advisability of the
introduction of the oxidant in the feed circuit of an NPP with the RBMK-1000 reactor is
not considered to be clearly proven.
The purpose of this paper is to evaluate, on the basis of the Soviet and foreign ex-
perience of operation of NPPs with boiling-water reactors and oxidant feed-up, both the
advantages of this regime when employed in the NPP with RBMK-1000 reactor and possible
complications which its introduction creates for the main structural materials in the active
zone and CCL (where the repair work is most difficult).
It can be said a priori, that the oxidant introduction in the condensate circuit will
not change the water chemistry indices of the CCL. The dosage of the oxidant in the feed
circuit will increase the oxygen concentration in the CCL, along with other factors,, due
to water radiolysis, which can, in turn, decrease the reliability of operation of channels
and lines made of 08Kh18N10T steel. Oxygen dosage into the condensate feeding circuit (CFC)
of the VK-50 reactor caused its concentration in the reactor water to increase from 180 to
250 ug/kg [7]. This method, therefore, deserves careful study.
Translated from Atomnaya Energiya, Vol. 59, No. 6, pp. 409-413, December, 1985. Orig-
inal article submitted April 5, 1985.
0038-531X/85/5906-0976$09.50
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
1L to 1vilV wil Val4 .. aG 111LLVU4l. l.1 V11 V1 L11G VA V 1G~', 1131G 111 1.11G 11G6V kJVWGl 11144.71~1y
caused unexpected failures at steam power plants (SPP) [8-11]: corrosion-erosion damage
of high-pressure heater (HPH) steam attemperator tubes on the heating steam side, damage
of the internal partitions of the HPH hot well and heating steam condensate pipeline,
failure of the convective HPH_coils made of Kh18N12T steel in the overheat zone, and clogging
of the turbine flow section with deposits.
The observed reduction in the iron ((Fe]) concentration in the reactor water from 22
to 10 }~g/kg during the VK-50 service is connected with the oxygen introduction in the feed
circuit [7, 12-14]. The change in the [Fe] in the reactor water during the whole period
of operation of the plant shows that the [Fe] decrease was observed at various times and
that no additional measures were taken. Thus, with no correction regime employed, in 1966-
1977 the [Fe] content in the rector water decreased from 75 to 22 ug/kg. .The general pat-
tern established for the [Fe] content in the reactor water, which decreased during various
periods of plant operation, can most probably be explained by kinetic factors rather than
just oxygen dosage (see Fig. 1).
In the VK-50 reactor 33% of the total area is accounted for by pearlitic steel parts,
and 14% by austenitic stainless steel. In the RBMK reactors the total area of parts made
of pearlitic steels does not exceed 5% of the total area of parts made of structural materials,
and in the CFC where the passivation of stainless steel is possible the total area of the
stainless steel parts is 1.2% and 50%. The remaining sections of parts made from structural
materials are already operating in oxygen-containing media (up to 7 ug/kg). If the oxygen
dosage has not changed the established pattern with regard to the iron balance in water
in the VK-50 reactor, the chance of this happening will be even less in the RBMK-1000 re-
actor where the oxygen introduced can passivate only 1.2% of the total area. The results
of the first stage of the oxygen dosage in the condensate circuit of the RBhIlt-1000 reactor
(the Third Unit of the Chernobyl'skaya NPP) have confirmed the forecasts made: neither
positive nor negative effect has been observed.
It should be noted that the [Fe] content in the reactor water of the VK-50 reactor, after
17 years of service including about four years of operation in oxygen regime (10 ug/kg)
is higher than the [Fe] content in the reactor water of the First Unit of the RBMK-1000
reactor after nine years of operation at the Leningradskaya NPP (2-7 ug/kg); this can be
explained, probably, by the ratio of the areas of pearlitic steel parts.
Consider the possible change in the [Fe] content in the feed water of the RBMK-1000
reactor with the oxygen dosage, by making several assumption, for simplicity. It is known
that various authors gave various evaluations of the effect of the oxidant introduction
in water; they pointed out both negative and positive effects on the corrosion velocity
in pearlitic steel, the estimated effect being 5-10-fold.
Assumption One. It is assumed that the oxygen introduction in the CFC of the NPP with
the RBMK-1000 reactor leads to the maximum, i.e. about 10-fold decrease in the pearlitic
steel corrosion.
Assumption Two. The interaction of the metal with water and the oxidant contained
in it will take part not at the already oxidized (as usual) surface but with the unoxidized
surface, the expected effect being maximum. Assume the amount of corrosion product removed
with water being equal for pearlitic steel to 50% of the corrosion rate value.
Assumption Three. The general corrosion rate of chromium-nickel austenitic steel in
water virtually does not change with the change in the oxidant concentration from 0.02 to
0.2 ug/kg, therefore, assume the steel corrosion rate to be equal to the corrosion rate
under the regime without correction. The amount of the corrosion product removed we assume
to be equal to 10% of. the corosion rate for the both cases (water with and without the
oxidant).
The areas of parts made of pearlitic steel: the feed circuit, 1500 mz; the condensate
circuit, 1300 mz; deaerators, 700 m2, the rest of the surfaces of parts made of pearlitic
class steel are already in contact with oxygen-containing steam flows. The total area of
parts made of austenitic steel in the CFC is 1800 m2.
Consider two possible variants of the oxidant dosage: its introduction both ahead
of and after the deaerator, and ahead of the deaerator only. In both variants deaerator
tanks (700 m2) are not subjected to passivation. To evaluate the change in the iron carry-
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
Fig. 1. Iron concentration in
the reactor water, VK-50 [7,
12-14]: ?) operation in the
regime without correction; X)
operation with the oxygen
dosage in the CFC (about 200
ug~kg)?
TABLE 1. Evaluation of Iron Carryover
to the Coolant [15, 16]
TABLE 2. Corrosion Products Carryover from Part Surfaces in the CFC of the NPP
with the RBI?II~ Reactors
over to the coolant, we will use the following known data [15, 16] tested. during operation of
units (see Table 1). Possible variations in the iron carryover to water in the CFC of the
NPP with the RBMK-1000 reactors are listed in Table Z.
In accordance with optimistic evaluations, when the oxygen dosage is being carried
out according to the first variant, with the account taken of the corrosion products formed
in the CCL (485T-0.5), the total amount of the corrosion products entering the CFC
(651.1T-0.5), and the coefficient of condensate purification equal to 0.5, the amount of
the corrosion products will be (651.6 + 485 + 263.2)T-0.5 = 1400T-0?y, while its observed
value in the water regime without correction is 1800T-0.5. Thus, the decrease in the amount
of the corrosion products entering the reactor will be ~ 207 of the total amount, of the
corrosion products. If the oxidant dosage is carried out ahead of the deaerator only, the
maximum possible positive effect will drop to 57. This means that the dose rate near the
CCL equipment will decrease by 207 at best, and may decrease just by 57. With the account
taken of the total error in the dose rate measurement, this positive effect, i.e. the de-
crease in the dose rate caused by the oxygen, can go unnoticed.
The decrease in the coolant radioactivity and. corrosion in the condensate circuit is
being reached at the Japanese reactors by combining the following measures [17]: increased
condensate purification, use of 'dry' and 'wet' conservation during prolonged and short
shutdowns, installation of additional filters ahead of condensate purification filters to ,
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20 :CIA-RDP10-021968000300070006-9
~+ b I ~N D\ N N ~
W W
Vi +.~ G
V1 ro O
I M V Y
?~ U ? ro ,-i C
~ .??. N \ O
v~
SO?+
3 ?~
o~ N , i ~ U
Q
-~
o
O
00
ro k *
O
q
~
U ~
+~ .C ~ U Vi
U
O N
~
b
O b
?,-i AA O UI
~
i~
N O
7i
U ro
C x ~ U 4+
U
o0 3
~, +~
x
cn v~
x
o ra
m
I ~
~+ o
U
U d ?'?I
~
U
V] +~ ?-i
O!]
H
w
a
~+bm
U ~ H
?~ro
~
~?>i
b
G
ro
~
ro N 1-I
~ ~ ~
b v~
~
v~
U
,~4
c
C
U T
-I
i
ro
V V~
V 3 nU
U~ U
t~A
U
V V
H U
H H
05 ~ 00
r-I H H
3-I
O
a
m
o
?~
d~
~
O
U
ro
OA 3-I
+~
F+ 00
ro
?'i
~, N
U
N O
d
b
k ~
~
ro
+~ r-i
~
i-I
ro
O ?,?i
~
U
4-I ro ro O
~ ~
3-+
~
O
~
~,.
.C
OA +-I
U
ro
~ .C
k
+~
Vi
?ri ~,
x
a~
F.,
?rl QO
?~
O
a
01.
~
M
b
G1
.L'
?ri
VI
~/ ?ri .~
f.,
~., is
d
N
k +?~
U b
O N
i?+
Vi
v1 3-I O
O G
?~ ~
Q
~
a
a -
z
~ ~ b
ro N N
N
O O O ,~G
~ +~ ~+ ro
+~
?~ ~,
o ro
a v, ca
d x +~ +~
i-I ~ U ?r?I b
~ ?~ o v, ~+
~ x x ?; .?
~ ~ ?~? a ~
E-+ w ~a ~ x
Declassified and Approved For Release 2013/02/20 :CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
trap nonsotuote corrosion proaucts, use in nearing systems of aus~eni~ic s~eel wiui i~wer
cobalt content, and introduction of oxygen in the CFC.
From this is follows that to decrease the amount of the corrosion products entering
the reactor it is necessary to provide the oxygen dosage in the feed circuit; the latter
will, in turn, cause the radiolysis increase in the CCL water, a further increase in the
oxygen concentration in the reactor water [7], and, therefore, the increase in the corro-
sion rate of zirconium alloys and will .increase the probability of the development of
corrosion cracking in parts made of 08Kh18N10T steel. The above factors can lead to a de-
crease in the service reliability of the reactor basic equipment.
It is appropriate to remind about failures of the basic equipment at the SPP [8-11]
and NPP [18-23] observed during operation with the oxygen dosage (see Table 3). Obvious
advantages of the employment of the oxygen regime in the NCC with RBMK-1000 reactors (as
compared with the regime without correction) are substantially decreased because of this
damage of equipment. Corrosion damage of equipment, technological problems, unavoidable
materials substitutes due to the oxidant use, increased deposit formation in the turbine,
use of various (in various units) oxidants (oxygen, hydrogen peroxide, air, etc.),, indicate
that no optimal water regime with the oxygen dosage is known today that can be used on the
SPP, even though it has been reported [27] that a weak alkali water regime (pH 8-8.5 NH,,OH)
with the oxygen dosage of up to 200 ug/kg can be considered today a rational regime for
the CFC of SPP power units.
At foreign NPPs with BWR reactors corrosion damage of the main equipment has been re-
corded systematically. In 1975-1980 the number of intercrystalline corrosion cracking cases
at NPPs all over the world increased from 64 to 213 [18-23]. More recent data (19, 23,
24] confirm a continuous increase in the number of cases of corrosion cracking in pipes
of BWR reactors (see Table 3). At Swedish NPPs operating in the oxygen water regime cases
were observed of intercrystalline corrosion cracking of stainless steel [25] which .made
it necessary to turn to the old water regime without correction. To lower the steel sensi-
tivity to corrosion cracking, hydrogen is introduced in the circuit [25, 26].
In the existing nuclear reactors restriction of the oxygen concentration in the coolant
proved to be an efficient measure. Recorded cases of corrosion damage in reactors with
increased oxygen concentration in the coolant confirm the fact that the oxygen regime has
not been mastered yet and give us concern about the oxygen dosage in the CFC in the NPP
with the RBMK-1000 reactors.
The increase in the oxygen concentration in the CCL (caused by the radiolysis in-
crease), provided the oxygen dosage in the feed circuit is employed, will make the opera-
tion of equipment made of stainless steel more complicated. It is known [28] that corro-
sion cracking of OKh18N10T steel is fixed in case of concentration of chloride-ions at steel
surfaces and is intensified with the increase in the oxygen concentration in the medium.
Particular attention should be given to increasing the oxygen concentration in the
CCL water as regards the operation of technological channels made of zirconium alloy with
2.5~ Nb, based on 30-yr reactor service life. Corrosion of such an alloy increases 4-fold
in 3500 h if the oxygen concentration in water is increased from 0.3-0.6 to 12-17 mg/kg
under radiation conditions [29].
The advantages and possible complications encountered when the oxygen dosage is used
in the CFC of NPP with the RBMK-1000 reactors can fie formulated as follows:
the oxygen dosage in the condensate circuit (200 ug/kg) can reduce the [Fe'] content
in the CCL by 57 at best, as compared with the value now defined;
the oxygen dosage in the condensate and feed circuits can reduce the (Fe] content in
the reactor water by about 207. At the same time, the corrosion rate of the base metal of
technological channels will increase, and corrosion of welded seams and adjoining zones
will increase with the inevitable decrease in the channel service life. The increase in
the oxygen concentration in the reactor water will increase the risk of the development
of corrosion cracking in type OKh18N10T steel in the most stressed sections of the CCL.
The above advantages of the oxygen dosage in the feed circuit by no means warrant the
worsening of operating conditions for stainless steel and zirconium alloy (with 2.5~ Nb)
causing the decrease in their serivice life in the CCL.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9 _
i.roicivi~;, ... a.... .. JB..,. .....~...g... ~.. ...... .,.,.,........,........ ..~~....~.. ..,.~) ...~ .. .... ......... ~.....~...... .......,~..
Bible for the NPP with RBMK-1000 reactors.
1. I. A. Varovin, A. P. Eperin, M. P. Umanets, and V. G. Shcherbina, "Ten-year experience
of operating the Leningrad NPP," At. Energ., 55 ,. No. 6, 349 (1983).
2. M. E. Shitsman, Yu. I. Timofeev, L. S. Midler, et a1., "27,000 h of the NRG operation
without serivice acid washing," Snergetik, No. 12, 4 (1978).
3. Yu. A. Egorov and I. Ya. Emel'yanov, "The state and problems of studying the radiation
safety at the NPP in connection with further development of nuclear power," in:
Radiation Safety and Protection at the NPP [in Russian], Energoizdat, 7, Mosocw (1982),
p. 5.
4. Yu. A. Egorov, A. A. Noskov, V. P. Sklyarov, et al., "The study and use of the TRAKT-I
model for calculation of the corrosion products activity in the technological circuit
of the NPP with a channel-type reactor," ibid., Atomizdat, 5, Moscow (1981), p. 5.
5. V. 5. Grechishkin, Yu. A. Egorov, G. N. Krasnozhen, et al., "Radiation Conditions at
the Chernobyl'skaya NPP in the Initial Operation Period," ibid., Energoizdat, 7, Moscow
(1982), p. 92.
6. N. I. Bogdanov, A. V. Borunova, Yu. A. Egorov, et al., "Corrosion Products in the CCL
(Controlled Circulation Loop) of the NPP with the RBMK Reactor," ibid., Energoizdat.,
8, Moscow (1984), p. 22.
7. E. P. Anan'ev, A. B. Andreeva, I. S. Dubrovskii, et al., "Efficiency of using the neu-
tral-oxygen water chemistry regime in operating the NPP boiling shell-type reactor,"
At. Energ., 52, No. 1, 10-14 (1982).
8. N. I. Gruzdev, Z. V. Deeva, B. E. Shkol'nikova, et al., "Possibility of the development
of brittle fractures in the heat boiler surface in the neutral-oxidizing regime,"
Teploe'nergetika, No. 7, 8 (1983).
9. V. I:`Gorin, "Some results of operating power units at supercritical pressure in neu-
tral-oxidizing water regime," ibid., p. 2.
10. N. A. Lyashevich, "Operation reliability of heat surfaces of power units in the water
regime with the oxidant dosage," ibid., p. 11.
11. G. P. Sutotskii, G. V. Vasilenko, Yu. V. Zenkevich, et al., "Water chemistry regimes
of SCP (supercritical pressure) units," in: Water Treatment and Water Chemistry and
Corrosion of the SPP and NPP [in Russian], TsKTI (Tsentr. NIiPKkotloturbinnyi Inst.),
158, Leningrad (1978), p. 20.
12. A. I. Zabelin, A. B. Andreeva, Yu. V. Chechetkin, et al., "Corrosion and Activation
in the Circuit of the NPP with a Boiling Reactor in Neutral Water Chemistry Regime
. without Correction," Preprint 364 [in Russian], NIIAR-5, Dimitrovgrad (1979), pp. 1-14.
13. A. I. Zabelin, A. B. Andreeva, V. M. Eshcherkin, et al., "Use of carbon steel in the
water chemistry regime without correction of the NPP VK-50," At. Energ., 49, No. 4,
229-232 (1980).
14. A. I. Zabelin, "Study of Water Chemistry Regimes of the NPP VK-50," Preprint 538
[in R.ussian], NIIAR-23, Dimitrovgrad (1982).
15. V. V. Gerasimov, Steel Corrosion in Neutral Water Media [in Russian], Metallurgiya,
Moscow (1981).
16. V. V. Gerasimov, Corrosion of Reactor Materials [in Russian], Atomizdat, Moscow (1981).
17. J. Mushima, "Water chemistry in the Japanese light water reactors," in: Proc. IAEA
Specialists Meetings on Influence of Power Reactor Water Chemiitry on Fuel Cladding
Reliability. Italy, Pisa, 12-16 Oct. 1981. Vienna: IAEA, 1982, p. 230.
18. J. Danko and K. Stahlkopf, "An overview of boiling water reactor pipe cracking," J.
Pressure Vessels and Piping, No. 9, 401-419 (1981).
19. C. Cheng, "Intergranular stress-assisted corrosion cracking of austenitic alloys in
water-cooled nuclear reactors," J. Nucl. Mater., 57, No. 11, (1975).
20. W. Casto, "Recent occurrences at nuclear reactors and their causes," Nucl. Safety,
15, No. 4, 466-477 (1974).
21. W. Casto, "Recent occurrences at nuclear reactors and their causes," ibid., No. 6,
742-750.
22. "Pipe cracking in boiling water reactors," Nuclear Regulatory Commission report, Nucl.
Safety, 17, No. 4, 475 (1976).
'L3. D. Locke, "Review of experience with water reactor fuels 1968-1973," Nucl. Eng. De-
sign, 33, No. 2, 94 (1975).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
24. M. Taylor, "Boiling water reactor stress corrosion cracking of piping-utility industry
research program," ibid., 69, No. 2, 223-227 (1982).
25. "Hydrogen stops growing cracks in BWRs," Nucl. Eng. Int., 25, No. 295, 6(1980).
26. G. M, Kalinin, "Intercrystalline corrosion cracking and methods of recovering NPP pipe-
lines," At. Tekh. Rubezhom, No. 1, 17-20 (1985).
27. O. .I. Martynova, "Further development of oxidizing water chemistry regime at power
units with direct-flow channels in Western Europe," Energokhozyaistvo Rubezhom, No.
3, 8-13 (1982).
28. "Structure materials corrosion under operation regimes of the first loop of the NPP
with boiling reactors. Stainless, and pearlitic steels and zirconium alloys," in:
Proc. of the Jubilee Conf. on the Occasion of the 20th Anniversary of Nuclear Power
Engineering [in Russian], Obninsk, June 25-27 (1974), p. 201.
29. A. S. Zaimovskii, A. V. Nikulina, and N. G. Reshetnikov, Zirconium Alloys in Nuclear
Power Engineering [in Russian], Energoizdat, Moscow (1981).
A. M. Rozen, A. S. Nikiforov, UDC 66.061.5:546.7
Z. I. Nikolotova, and N. A. Kartesheva
Previous studies [1-12] have established rules linking the extractive ability of mono-
dentate organic compounds with their structure (electronegativity of the substituents X,
electron density at the reaction center q, alkalinity pKa, et al.). It was found that the
extractive ability increases with q on the functional atom of the extractant (and corre-
spondingly with pK) and decreases .when the electronegativity of the substituents increases,
since ~q~ = a" - b"EX (Fig. 1),
1gK=A-BEX=a'-I-b' ~ 4 ~ =a-I-bPKa? (1)
It was also established that the length of the hydrocarbon chain has virtually no effect
on extraction: curves with a weak maximum when the number of carbon atoms NC = 8, are ob-
served (Fig. 2).
In a different series of studies the extractive ability of bidentate compounds [12-20]
and crown esters [21] was studied.
The rules found enable controlling the extractive ability. They were used in subse-
quent studies and in the selection of extractants (mono-, bi-, and polydentate) for extract-
ing-actinides. ,
Monodentate Extractants
We posed the problem of improving the extraction system based on TBP, widely used all
over the world for regenerating spent nuclear fuel. The extractive properties of TBP are
practically optimal, but physically they are not entirely satisfactory: the solubility
in the water phase is too high, the solvates of quadrivalent actinides are poorly compatible
with the hydrocarbon diluents (a secondary organic phase already forms at moderate concentra-
tions of thorium and plutonium nitrates [22, 23]).~ Efforts to improve this system were
made both in the USSR [24-27] and in the USA [28]. We posed the problem of improving the
physical properties of the extractant, while preserving the extractive ability based on
TBP. The latter condition, as follows from theoretical considerations [1-12], requires
preserving in molecules of neutral phosphoorganic extractants (NPOE) the same values of
the electronegativity of the substituents as in TBP, i.e., the use of compounds of the same
class - trialkylphosphates, since a change in the chemical nature of the substituent gives
rise to a significant change in the electronegativity and, as a consequence, in the extrac-
tive ability (see Eq. (1)). The physicochemical properties can be improved by optimizing
the hydrocarbon chain: reducing the solubility of esters in water by lengthening the chain
and improving the compatibility of the solvates with long-chain hydrocarbon diluents by
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 413-421, December, 1985. Orig-
inal article submitted March 25, 1985.
0038-531X/85/5906-0982$09.50 ? 1986 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
n. O ~ y.
m
~ p A H R,, N+ AS NH' RZNNQ RN H3
-2 -7 0 7 2. 3 4 p/(Hep
~ i t i
's------~~------ - ~5- --PN,NM
Fig. 1. Dependence of the extractive ability of organic compounds
relative to the actinides and acids on the structural properties of
the extractants: a) neutral phosphoorganic compounds (the series
TBP-TOPO); b) amines (series of primary, secondary, and ternary
amines, quaternary ammonium bases); c) series of neutral organic com-
pounds TBP (tributyl phosphate)-DAMF (diisoamylmethyl phosphonate)-
TOPO (trioctyl phosphene oxide)-TOASO (trioctyl arsenoxide)-TOAD
(trioctyl amine oxide); the numbers in parentheses indicate the
number of molecules of the organic ligand iri the complex; c~PO is the
frequency of the stretching vibrations of the PO group; EX is the
sum of the electronegativities of the substituent groups (XR = 2.0;
Xg0 = 2.9; Xg = 2.3); Eo* is the sum of the Taft constants for the
substituent groups; pKgZO and pK~ are the alkalinity of the organic
compounds referred to water and nitromethane.
increasing the length of the hydrocarbon chain of esters and using isostructures. As indi-
cated above, these structural changes have virtually no effect on the extractive ability.
The theoretical forecast was confirmed experimentally [27]. Because of the difficulties
of flushing the acidic impurities (products of hydrolysis and radiohydrolysis) in the case
when the length of the hydrocarbon chain is increased, trialkylphosphates with 1.n~ = 15-18
[24-27], both symmetrical [(i-CnHzn+10)3P0, where n is equal to 5 or 6] and heteroradical
(for example, diisobutylisooctyl phosphate, End = 16), were recommended. .
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
S ~ ~`.
~x
..
4
3
~~
7 B 9 10 11 17 n~
Fig. 2. Dependence of the extraction
constants of uranyl and plutonium
nitrates (---) on the length of the
hydrocarbon chain of the extractant:
x) amines; ?) phosphonates..
We note that in the USA triisoamyl phosphate (EnC = 15), trihexyl phosphate (EnC = 18),
and triisooctyl phosphate (EnC =.24) were recommended. A successful check of the latter in
hot chambers was reported in [28]. A report of the effective flushing of long-chain acidic
products of hydrolysis, i.e., D2GFK, by the usual methods is puzzling.
Bidentate Extractants
The second practical problem is the selection of effective extractants for extracting
transplutonium elements and lanthanides from the discarded solutions of radiochemical tech-
nology. As pointed out previously [15, 20], because of the so-called effect of anomalous
aryl strengthening (AAS), tetraaryl methylene disphosphonine dioxides have the highest ex-
tractive ability relative to trivalent actinides and lanthanides, permitting their extrac-
tion from solutions with any acidity without preparation: However, they are poorly compat-
ible with hydrocarbon diluents; compatibility can be improved by introducing alkyl radicals
into the benzene cores, stabilization (for example, with tributyl phosphate [15, 29]), as
well as changing over to mixed aryl-alkyl dioxides, which makes the synthesis more compli-
cated.
In recent years interest has appeared in bidentate compounds with the groups P = 0
and C = 0 - dialkyldialkyl carbamoyl methylene phosphonate and phosphine oxides, which are
undoubtedly easier to synthesize than dioxides, and their compatibility with hydrocarbon
diluents is appreciably higher [29-34]. It is therefore desirable to study the change in
the extractive ability in a wider range of bidentate phosphoorganic compounds (from diphos-
phine dioxides to carbamoyl phosphine oxides and phosphonates), to discuss the controversial
questions of the coordination of actinides accompanying the use of carbamoyl compounds,
as well as to establish whether or not the AAS effect exists in these systems (according
to [32) it does not exist and according to [33J it does exist). Of course, since the effect
was observed during extraction of nonsymmetrical diphosphine dioxides, one would expect
that it also exists in the case of carbamoyl phosphine oxides.
Effect of Anomalous Aryl Strengthening. As is evident from Eq. (1), the introduction
of electronegative substituents into the molecule of the extractant decreases the donor
ability of the functional atom (on our case oxygen) and lowers the alkalinity and extractive
ability. The only breakdown of this rule was observed in extraction by dibentate phosphoor-
ganic compounds - diphsophine dioxides RZP(0)CHZ(0)PR2. When electronegative groups such
as RO (X = 2.4), C1(CH2)2 (X = 2.36) were introduced, the rules (1) were obeyed. when the
alkyl substituents (X = 2) were replaced with more electronegative phenyl groups (Xph =
2.36) the alkalinity, .as expected, decreased; judging from the changes in the infrared spec-
tra (increase in the frequency of the stretching vibrations of the group P = 0) and in the
extraction of HN03 (characterized by monodentate coordination), the charge on the oxygen
and its donor and the extractive ability decreased. The extraction of trivalent actinides
and lanthanides nevertheless increased substantially (Figs. 3a, b, c) [13, 14]. Since the
same effect was also observed when other aryl substituents were introduced, it was called
AAS. It was-found that it remains when the nitrate medium is replaced by HC1, HzSO,, and
is especially large in a HC10,, medium.
It turned out, however, that the effect vanishes when the aryl substituent is separated
from the phosphorus by the CHz group, which interferes with the conjugation (for example,
when the phenyl is replaced by benzene), and when the bridge is lengthened - when the
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
it o a i
[g K
10
Fig. 3. AAS effect for complexes accompany-
ing extraction of trivalent actinides and
lanthanides by diphosphine dioxides (0.01
mole/liter solutions in dichloroethane): a)
distribution factors [1) (C6H5)tP(0)CHz(0)P(C6Hs)t
(or 4 Ph); 2) (C6H5)ZP(0)CHZOP(C8H17) (or 2 Ph
2 oct); 3) (C8H17)ZP(0)CHZOP(CgHl~)Z (or 4 oct);
4) (C,,H9)tP(0)CHt(0)P(C,,H9)t (or 4 But); 5)
TOPO (0.2 mole/1); 6) TOPO (0.2 mole/1)]; b)
effect of the length of the alkyl bridge
I.1) 4 Ph; 7} (C6H5)tP(0)(CHt)t(0)P(C6H5)tS 8)
(C6H5)tP(0)(CHt)3(0)P(C6H5)]; c, d) dependence of
the extraction constants on the.electronega-
tivity sum EX and alkalinity pKa (- dichloro-
ethane diluent, ---- chloroform diluent).
methylene bridge is replaced by an ethylene or propylene bridge (Fig. 3d). The replacement
of the ethylene bridge by a vinylene bridge CH=CH restored the effect. Calorimetric measure-,
ments show that the effect has a bonding nature (the enthalpy of extraction of europium
by tetraphenyl dioxide PhtP(0)CHt(0)PPlit was equal to 11.6 kcal/mole, as compared with 8.8
kcal/mole for tetraoctyl dioxide). All these facts can be explained by assuming that com-
plexing is accompanied by delocalization of the electron density from the aryl groups into
the central cycle of the complex and, possibly, aromatization of the six-term or, in the
case of dioxides with the vinylene bridge (CH = CH), even the seven-term cycle formed. The
latter proposition is supported by the high mobility of the protons in the methylene bridge
in the complex observed in [19]. They become capable of isotropic exchange with chloroform
(while the proteins of the ligand bridge are not capable of exchange).
Study of the Extractive Ability of Bidentate PhosphoorQenic Compounds. We studied
the following compounds: carbamoyl methylene phosphonates (i-C8H170)ZP(0)CHZC(0)N(CZHS)2
[or (i-oct0)t/Ett], (i-CSH110)ZP(0)CHZC(0)N(C4H9)2 [or (i-amyl 0)t/Butt], phosphine oxides
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
~Am x~"\
x
10~ . ~.
~i/ "~ \x~,~,x~x
+/`t'_+\
+~+~
2 J 4 5
[H.NOslwater~ mole/liter
~ i i i 1 1
4 6 B 10 12 J4
(HNOjlwater~ mole/liter
Fig. 4 Fig. 5
Fig. 4. Extraction isotherms of HN03 extracted by bidentate phosphoor-
ganic compounds (the diluent is dichloroethane): z = yg/Lo, 0) 4 oct;
o) 4 Ph; +) octt/Butt; O) amyl OZ/Butt; ---) TOPO.
Fig. 5, Effect of the structure of bidentate phosphoroorganic compounds
on their extraction ability (0.05 mole/liter solutions in dichloroethane);
x) 4 Ph; O) 2 Ph 2 oct; ~) Tolt/Butt; ^) Pht/Butt; 0) octt/Butt; +)
(i-amyl 0)t/Butt.
(C8H17)ZP(0)CHZC(0)N(C,,H9)2 (or octZ/Butt), (CsH;)tP(0)CH2C(0)N(C,,H9)t (or Pht/Butt),
(C6H,,CH3)tP(0)CHZC(0)N(C,;H9)t (or Tolt/Butt), dioxides (C6H5)tP(0)CHZP(0)(CRHi~)t (or 2 Ph
2 oct) and (C6H5)tP(0)CHZP(0)C6H5)t (or 4 Ph).
' We extracted traces of americium acid other actinides from nitrate media, when the con-
centration of "free" extractant (L), which must be known in order to describe the equilibrium,
was determined by the extraction of HN03, which we studied up to concentrations of 13 mole/
liter (Fig. 4). The quantity z [HN03]org/Lo, where Lo is. the starting concentration of
the ligand (extractant), characterizes the number of HN03 molecules per ligand molecule.
As is evident from Fig. 4, the value of z for carbamoyl phosphonates approaches 4 (and there
is no saturation), i.e:, the complexes (HN03)i(Ht0)hL, where i ~ 4, form. Assuming that
HN03 molecules attach directly to both reaction centers (hl = ht = 0), that subsequent
molecules attach through water, and that h3 1, h~, = 2, we conclude that the complexes
HN03?L, (HN03)t?L, (HN03)3(Ht0)?L and (HN03)4(Ht0)?L form. From-the law of mass action
we find their concentrations yi in the organic phase:
yi = KidL; yi/a = K,/sa2L; yi/3 = Kilsa3ax.oL;
where the activity of the acids a = [HN03]waterY~ (Y? is the activity coefficient; aHtp is
the activity of. water). The total concentration of acid in the organic phase is given by
yx=F~-I-2y1/z+3y~/s-I-4y~/y -F- ..., (3)
while-the concentration of the free ligand is given by
L = L0/ ~1 '+' KiQ'f' KSI2aZ ~" K113a3ati8U -~- KiJ4a4aA80 -{"" ... ).
The least-squares determinations of the extraction constants are presented in Table 1, and
the extraction isotherms, calculated from Eqs. (2)-(4) and these values of the constants,
are presented in Fig. 4 (solid lines). The predicted values are in good agreement with
the experimental values.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
1HDLP~ 1. G]SL1Ci l:11V11 l.rV11J LGll 1.D Vl
Phosphoryl Compounds
Compound
Diluent
(i-oct0)z/Et;~
C2H4C12
(i-flmy10)2/Blltg
C2H4C12
oct2/Butt
C2H4C12
Ph2/Butt
C2H4C12
Ph2/Butt
CHC13
To12/Butt
C2H4C12
To12/Butt
CHCIa
2Ph 2oct
C2H4C12
2Ph 2oct
CHC13
4Ph
C2H4C12
4Ph
CHC13
xl I
Ki~2 I
`x1j3 I
K1j4 I
R2 I
x3
0,25
2,6.10-3
2,1.10-8
1,0.10-13
5,0.102
5.0.102
0,23
1,1.10-3
2,2.10-8
3,2.10-e
-
1,8.1x''
7,2
0,15
1,0.10-4
1,4.10-7
2,5.105
1,2.10'
0
89
1,3.10-2
8,8.10-5
1,6.10-8
3,8.103
2,5.107
,
0,16
5;6.10-4
2,1.10-8
1,0.10-8
1,5.102
1,3.104
1,7
4,8.10-2
2,0.10-4
1,0.10-7
1,0.108
2,0.107
0
23
1,3.10-3
5,0.10-8
1,0.10-11
1,0.103
1,0.102
,
2
4
0,33
-
-
1,4.108
1,3.102
,
62
0
4.10-2
1
2
4.10-5
1,0.10-7
57
1,3.107
,
0,85
,
7,0.10-2
,
-
-
7,0.108
8,0?l0io
0,14
4,0.10-3
2,8.10-8
8,4.10-8
8,0.105
1,6.108
The dependence of the distribution ratios of americium a~ on the acidity of the water
phase (xH) is shown in Figs. 5-8. In the experiments we used solutions of dioxides with a
concentration of 0.01 mole/liter, phosphine oxides with a concentration of 0.05 mole/liter,
and phosphonates with a concentration of 0.5 mole/liter. In Figs. 5-8 the data are scaled
to the concentration 0.05 mole/liter, in the case of dichloroethane as the diluent (apparent
solvation number equal to two) we multiplied the data for dioxides by 52 = 25, we divided
the data for phosphonates by 102, and for .dilution with chloroform (q = 3) we multiplied
the data for dioxides by 53 = 125. The curves in the figures just as those obtained in
[13, 14], have a maximum, characteristic for extraction of metals by the solvation mechanism
[in the form Am(N03)3Lq] and is determined by the combination of the salting out and dis-
placing action of HN03 (35]. The subsequent minimum-and growth of a~ are linked with the
changeover to extraction by the acid complexes HpAm(N03)3.}.pLq (probably p ~ 3), when a~
becomes proportional to (H+] with a high power of 3 + 2p. By the equilibrium shift method
(dilution) it was found that the apparent solvation number of americium is equal to 2.3
(dichloroethane is the diluent) and 3 (chloroform), i.e., di- and trisolvates are formed.
Correspondingly,
aAm = a2 m'+ a3 m = ~K2.AmLz ~" K3AmL3) [N~3~3 'Y?. (5 )
The values of the formation constants of di- and trisolvates KZ and K3, found from the data
in Figs. 5-8 by the least-squares method using Eq. (14), are presented in Table 1:
From the data in Figs. 5-8 and Table 1 we can draw the following conclusions:.
- extraction decreases in the series 4Ph > 2Ph 2oct > To12/Butt > Ph2/Butt >
oct2/Butt > (i-amyl 0)2/Butt (i-oct 0)2/Et2, independently of the nature of the diluent;
- the changeover from diphosphine dioxides to carbamoyl phosphine oxides lowers the
extraction of americium approximately by 3-3.5 orders of magnitude. We note that when one.
of the groups P = 0 was replaced by S = 0, a drop by only a factor of 500 was observed [36],
so that the replacement of the P = 0 group by C = 0 group lowers the extraction more strongly
than the replacement of P = 0 by S = 0. This is linked to the weaker extractive ability
of the group C = 0 (the amides RC(0)NRZ extract americium at the level of phosphonates).
Nevertheless the extractive ability of carbamoyl phosphine oxides remains very high (K3~
106 instead of 103 for the usual phosphine oxide);
- the AAS effect is observed in the extraction of americium by carbamoyl methylene
phosphine oxides just as for diphosphine dioxides (the series 4Ph - 2Ph 2oct - 4oct; see
Figs. 3 and 5): Ph2/Butt extracts more strongly than oct2/Butt (in this respect Ph2/Butt
is the analog of the dioxide 2Ph Zoct). Thus the results of [33] are valid, while the data
on [29] were not confirmed;
- extraction is appreciably increased by replacing phenyl substituents with tolyl sub-
stituents (see Figs. 5 and 6), which indicates that the AAS effect remains and that the
alkyl radical introduces an additional electronic effect;
- it is evident that the repalcement of alkyl substituents by more electronegative
alkoxy groups, i.e., the changeover from phosphine oxides to phosphonates lowers the ex-
traction constant by approximately a factor of 103. This result is trivial, and is in
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
2
0
y s a ~o ~2 ~~
[HN031water~ mole/liter
Fig. 6. Effects of diluents on the extrac-
tion of HN03 (a) (0.1 mole/liter of the
ligand) and Am (b) (0.05 mole/.liter of the
ligand) by carbamoyl phosphine oxides
Ph2/Butt (-) and To12/Butt (---): O )
dichloroethane; ~) C6H6; o) CC14; x)CHC13;
?) Ph2/Butt in isoamyl alcohol; ~ )
Ph2/Butt in C6H6 + 1 mole/liter TBP.
agreement with the previously obtained data for monoentate compounds [1-12] and for diphos-
phine dioxides as well as with the results of [32]; and,
- it was found that for carbamoyl compounds extraction increases in the series Am <
U < Pu.
We shall now discuss the questions of coordination. It was previously established
[13-20] that HN03 coordinates to diphosphine dioxides in a monodentate manner and to actinides
in a bidentate manner. In the case of carbamoyl phosphine oxides the nature of the coordina-
tion is undoubtedly the same: the preservation of the AAS effect indicates bidentate coordina-
tion of the actinides (to the phosphoryl and carbonyl oxygen atoms of phosphine oxide).
In [31] and elsewhere, however, an attempt was made to prove that for extraction with carbo-
moyl phosphonates the coordination is monodentate, as in the. case of the usual monodentate
phosphonates (RO)ZRPO, while the role of the second center reduces to creating barriers
to the extraction of HN03 ("buffer action"). Because of .this, a smaller fraction of the
extractint is bound by the acid and the distribution ratios of americium increase. As addi-
tional proof, it was pointed out that in the extraction of americium from LiN03 solutions
in the absence of HN03 the difference between its extraction by bidentate carbamoyl phospho-
nates and the usual phosphonates decreased significantly. These arguments, however, are
incorrect, because they ignore the fact that the CHZC(0)NRZ extracted less HN03 than the
monodentate analog RZP(0)CHZ(0)PR2; the extraction increased with the length of the alkylene
bridge, i.e., as the influence of the second center was weakened. And, without the coor-
dination by the center C = 0, in the lithium system americium would be extracted less
strongly by carbamoyl phosphonates than by monodentate phosphonates. In reality, the oppo-
site situation occurs. Thus the data in [31] prove not the point of view of the authors,
but rather the presence of bidentate coordination.
Effect of diluents. As is evident from Figs. 5-8, the nature of the diluents has an
unusually strong effect on the. extraction of actinides by bidentate phosphoorganic compounds,
expecially diphosphine dioxides and diaryldialkyl carbamoyl phosphine oxides. The distribu-
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
I unm ~of'~
,'Oz ~ \
X
~
~x~x/
I
10 ~
I
102
0 2 4 6 8 10 12 14~
[HN031water~ mole/liter
Fig. 7
[HNOjJ.wayer? mole/
-liter
Fig. 8
Fig. 7. Effect of the structure of bidentate phos-
phoroorganic compounds on their extractive ability
(0.05 mole/liter solutions in chloroform): X) 4Ph;
O) ZPh2oct; ~) To12/Butt; ^) Ph2/Butt.
Fig. 8. Effect of diluents on the extraction of
americium by TOPO and carbomoyl methyl phosphonates
(i-amyl 0)2/Butt; X).0.1 mole/liter TOPO - dichloro-
ethane; p) 0.1 mole/liter TOPO - CoH6; ?) (i-amyl 0)Z/
Bute - dichloroethane;.o) (i-amyl 0)2/Butt - C6H6.
tion ratios accompanying the use of dichloroethane and benzene differ by approximately a
factor of 200. In the meantime, in the extraction of HN03 (i.e., with monodentate coordina-
tion), the effect of the diluent is small (see Fig. 6). An exception is chloroform, which
forms H bonds with the oxygen in ligands.
The strong effect of the diluent on the extraction of actinides is explained by the
apppearance of an interaction between the diluent and complexes with bidentate coordination.
The mechanism of this interaction is still not understood.
It is evident from Figs. 5 and 6 that aryl carbamoyl phosphine oxides have definite
advantages, and especially ditolyl dibutyl carbamoyl methylene phosphine oxides To12/Butt,
whose extractive ability is increased by the AAS effect and the electronic effect of the
methyl radical. In addition, the introduction of hydrocarbon radicals and a benzene core
in accordance with the recommendations of [15, 37) lowers the compatibility with hydrocarbon
diluents, solubilization of TBP gives an additional effect (15, 29] (for example, when 1
mole/liter TBP is introduced, 0.5 mole/liter of the ligand solution can be used). Compounds
of this class are of definite practical interest and can be recommended for technological
studies.
Extraction by Polydentate Extractants (Crown Esters)
The use of stereo specific macrocyclical extractants can help to solve the problem
of selectivity, since this is precisely the way this problem is solved in nature (for example,
heme transfers only iron). Until recently the extractive properties of crown esters were
studied predominantly for alkali and alkaline-earth metals. The significant selectivity
of these extractants was noted. For example, crown esters of the type 18-crown-6 (the
first number is the number of atoms in the cycle and the second number is the number of
oxygen atoms) selectively extract potassium and strontium; 15-crown-5 selectively extract
sodium; crown-4 selectively extract lithium [38). These results were explained by the
structural correspondence between the ion sizes and the dimensions of the cavity in the
macrocycle.
Extraction of actinides was first studied by B. N. Laskorin et al. [39]. It was noted
that quadrivalent actinides, forming disolvates, are extracted strongly while hexavalent
actinides (monosolvates) are extracted weakly;. the values of the extraction constants were
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
^` ~\0/\I / ~\o/~
~a o~ `o
~~~ ~~~
a b
Fig 9 Fig. 10
Fig. 9. Structure of dibenzo-l8-crown-6 (a), 18-crown-6 (b), and dicyclo-
hexyl-l8-crown-6 (c).
Fig. 10. Extraction isotherms for HN03 extracted by 0.1 mole/liter solu-
tions of crown esters in dichloroethane: ~) dicyclohexyl-l8-crown-6, O )
18-crown-6; e) dibenzo-l8-crown-6; ---) TOFO, -?-) TBP; z = [HN03)org/
[L]start~ xH = [~Oslwater~ YH = [HN03]org?
0 Z0~ ~~~
XMe,g/liter
aPu
~~
~' ~0 2 4 6 B 10 12
xHNa3, mole/liter
Fig. 11 Fig. 12
Fig. 11. Extraction isotherms for Th(e) and U (O)
extracted by dicyclohexyl-l8-crown-6: yMe = [Me]org.
Fig. 12. Dependence of the distribution factors of
Pu (IV) on the HN03 concentration of 0.1 mole/liter
during extraction by solutions of crown esters in
dichloroethane: ?) dicyclohexyl-l8-crown-6; O )
18-crown-6; a ) dibenzo-l8-crown-6.
presented. In our studies [21] data are presented on the distribution of nitric acid and
thorium, uranyl, and plutonium nitrates in the extraction by several esters, predominantly
dibenzo-l8-crown-6, 18-crown-6, and dicyclohexyl-l8-crown-6 (Fig. 9). The chemical nature
was discussed and the quantitative characteristics of the extractive equilibria were found.
The very strong extraction of HN03, exceeding the extraction observed with the use of a
strong extractant such as TOPO, was most unexpected (Fig. 10). The mechanism of the extrac-
tion is not completely understood. In [21] we proposed to form the complexes (HN03)iHzOhiL
according to the type of damped chain polymerization with the group H30+ in the cavity and .
the groups N03- and H30+ or HN03 alternating in the perpendicular direction. But NMR studies
did not prove the coordination of two water molecules to each oxygen of the crown ester.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
~X
I j/I
os
~l I ~ ~ ,.,,
l
Q3~I~HNU,,I b~l~~~
2 3
-d GZ9e, kcal/mole
Fig. 13. Dependence of the ex-
traction constants of HN03 and
Th on the alkalinity of the
,crown esters: 1) dicyclohexyl-
18-crown-6; 2) 18-crown-6; 3)
dibenzo-l8-crown-6; OG298)
Gibbs energy of interaction
of the extractant-with phenol
[40l?
The extraction isotherms of thorium are not entirely the usual isotherms (Fig. 11). How-
ever, the observed dependence of the distribution ratios of plutonium on the acidity (curves
with a maximum in Fig. 12) is characteristic for extraction with neutral compounds [35]:
the acid first acts as a salting out agent and then as a competitor, binding the ligarid.
The previously established (for neutral organic compounds) dependence of the extractive
ability on the electronegativity of the substituents and on the alkalinity of the esters
was also observed: when electron-donor cyclohexyl substituents are introduced the extrac-
tion decreases (dibenzo-l8-crown-6) (see Figs. 10-12). The correlation with the alkalinity
is approximately linear (Fig. 13), though deviations from linearity are possible owing to
"favorable" conformations. This indicates that both spatial factors and to a significant
extent electronic factors play a role in extraction by crown esters, which can be explained
by the significant contribution of acceptor-donor interaction. We recall that alkali and
alkaline-earth elements interact with crown esters predominantly electrostatically. Acti-
nides, on the other hand, exhibit distinct acceptor properties, which is what gives rise
to the increase in the role of this interaction, and attenuates the role of structural corre-
spondence.
We shall now study the question of the application of crown esters for extraction and
separation of actinides. The strong extraction of plutonium and weak extraction of uranium
enables in principle the use of crown esters for separation of these elements. This is
hardly desirable, however, because amine salts. - which are incomparably cheaper and more
accessible extractants - have an analogous, though weaker, property.
Crown esters of the 18-6 type cannot, unfortunately, be used for both extraction and
separation of trivalent actinides and lanthanides - the alkalinity and extractive ability
of these extractants are inadequate. For example, in extracting americium with a 0.5 mole/
liter solution of DTsG-18-crown-6 in dichloroethane and the introduction of a salting out
agent [2 mole/liter A1(N03)3J, a distribution factor of only 0.05 was achieved.
The extractive ability of esters of the dibenzo-l8-crown-6 type, containing the phos-
phoryl group P = 0, was also checked. It turned out to be very low (approximately at the
level of the corresponding phosphonate), i.e., the combined (P = 0 and crown-ester) coordi-
nation did not arise.
It is desirable to develop macrocyclical extractants, whose spatial characteristics
and alkalinity are adapted to the extraction and fine separation of trivalent actinides
and lanthanides.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
i,?u., uc~~,a~~c riv~,ic~a ttao vccu autttcvcu 11t 111vcJl,i~j'dL1Vi1S d11U Se1eCL10I1 OS neW eX-
tractants for extracting actinides. However, although a series of mono- and bidentate com-
pounds can be recommended for practical applications, macrocyclical extractants for ex-
traction and separation of trivalent actinides require additional development.
LITERATURE CITED
1. A. M. Rozen and Z. I. Nikolotova, "Dependence of the extractive ability of organic
compounds on their structure and electronegativity of substituent groups," Zh. Neorg.
Khim., 9, No. 7, 1725-1743 (1964).
2. A. M. Rozen and N. A. Konstantinova, "Dependence of the extractive and reactive
abilities of organic compounds on their structure," Dokl. Akad. Nauk SSSR, 167, No.
1, 132-135 (1966).
3. A. M. Rozen, "Problems of physical chemistry of extraction," Radiokhimiya, 10, No.
3, 272-309 (1968).
4. A. M. Rozen, Z. I. Nikolotova, A. A. Vashman, et al., "Dependence of the extractive
ability on the structure of the extractant," in: Chemistry of Extraction Processes
[in Russian], Nauka, Moscow (1972), pp. 41-45.
5. A. M. Rozen, Z. I. Nikolotova, and N. A. Kartasheva, "Mechanism of extraction with
organic oxides P3X0 and bases P,,XN03 in the series N-P-As," Dokl. Akad. .Nauk SSSR,
209, No. 6, 1369-1372 (1973).
6. A. M. Rozen, Z. I. Nikolotova, and N. A. Kartasheva, "Some rules for extraction of
actinide elements," Radiokhimiya, 16, No. 5, 686-695 (1974).
7. A. M. Rozen, "Contr.ol of the extractive abiltiy of organic compounds," in: Hydro-
metallurgy [in Russian], Nauka, Moscow (1976), pp. 194-198.
8. A. M. Rozen and D. A. Denisov, "Approximate quantum-chemical justification of the equa-
tions of extractive ability (Hammet-Taft method of electronegativities)," Radiokhimiya,
18, No. 6, 921-923 (1976).
9. A. M. Rozen, Z. I. Nikolotova, and N. A. Kartasheva, "Effect of extractant structure
on extractive ability," Zh. Neorg. Khim., 24, No. 6, 1642-1651 (1979).
10. A. M. Rozen and A. S. Skotnikov, "Effect of the structure of compounds in the series
(RO)3P0 - R3P0 - R3A's0 - P3N0 on the extraction and nature of complexification with
HReO,, and HTcO,,," Dokl. Akad. Nauk SSSR, Z59, No. 4, 869 (1981).
11. A. M. Rozen, Z. I. Nikolotova, N. A. Kartasheva, et al., "Effect of the structure of
organic compounds on their extractive ability," Radiokhimiya, 25, No. 5, 603-608 (1983).
12. A. M. Rozen, "Dependence of the extractive ability on the structure of the extractant
and separation of the contributions of solvation and hydration to the equilibrium con-
stant," in: Extraction Chemistry [in Russian], Nauka, Novosibirsk (1984), pp. 68-95.
13. A. M. Rozen, Z. I. Nikolotova, N. A. Kartasheva, et al., "Extraction of americium by
diphosphonic dioxides," Radiokhimiya, 17, No. 2, 237-243 (1975).
14. A. M. Rozen, Z. I. Nikolotova, N. A. Kartasheva, et al., "Anomalous dependence of the
strength of americium (III) complexes and other Me (III) complexes iwth diphosphonic
dioxides on their structure," Dokl. Akad. Nauk SSSR, 222,.No. 5, 1151-1154 (1975).
15. A. M. Rozen, Z. I. Nikolotova, N. A. Kartasheva, et al., "Diphosphonic dioxides -
unique extractants for extraction of actinides," in: Abstracts of Reports at the 2nd
All-Union Conference on the Chemistry of Transplutonium Elements, Dezisy Dokladov,
Dimitrovgrad (1983), p. 10.
16. A. M. Rozen, Z. I. Nikolotova, and N. A. Kartasheva, "Anomalous aryl strengthening
of complexes in extraction of americium and europium by alkaline diphosphonic dioxides
from perchloric media," Radiokhimiya, 20, No. 5, 725-734 (1978).
17. A. M. Rozen, Z. A. Berkman, L. E. Bertina, et al., "Extraction of nitric acid by
alkaline diphosphonic dioxides," Radiokhimiya, 18, No. 4, 493-501 (1976).
18. A. M. Rozen, Z. I. Nikolotova, N. A. Kartasheva, et al., "Extraction of americium by
vinylene diphosphonic dioxides," Radiokhimiya, 18, No. 6, 846-847 (1976).
19. A. M. Rozen, V. V. Akhachinskii, N. A. Kartasheva, et al., "Reasons for the anomalous
aryl strengthening of actinide and lanthanide (III) complexes with diphosphonic
dioxides," Dokl. Akad. Nauk SSSR, 263, No. 4, 938-942 (1982).
20. A. M. Rozen, Z. I. Nikolotova, and N. A. Kartasheva, "Diphosphonic dioxides as ex-
tractants for actinide elements (in connection with the problem of anomalous aryl
strengthening of complexes)," in: Research in the Reprocessing of Irradiated Fuel
[in Russian], Atomic Energy Commission of Czechoslovakia, Vol. 2, Prague (1977), pp.
22-29.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
21. A. M. Rozen, G. 1. Nikolotova, N\ A. Kartasheva, et al., "~;xtraction oT actinides and
nitric acid by crown esters," Dokl. Akad. Nauk SSSR, 263, No. 5, 1165-1169 (1982).
22. A. Rozen, "Problems in the physical chemistry of solvent extraction," in: Solvent
Extraction, Proceedings of the International Conference, North-Holland, Amsterdam
(1967), pp. 195-235.
23. A. Mills and W. Logan, "Third phase formation between some actinide nitrates and 20~
tri-n-butylphosphate odourless kerosene," ibid., pp. 322-326.
24. A. M. Rozen, A. S. Nikiforov, V. S. Shmidt, et al., "Method for extracting actinides,"
Inventor's Certificate No. 841140, Byull. Izobr., No. 14, 319 (1982).
25. A. S. Nikiforov, V. S. Shmidt, and A. M. Rozen, "Choice of organic solvent for ex-
traction processes in regeneration of spent nuclear fuel," in: Abstracts of Reports
at the 13th Mendeleev Conference, Nauka, Moscow (1981), p. 182.
26. A. S. Nikiforov, V. S. Shmidt, A. M. Rozen, et al., "Physicochemical foundations for
the selection of organic solvent for extraction processes in regeneration of spent
nuclear .fuel," Radiokhimiya, 24, No. 5, 631-636 (1982).
27. A. M. Rozen, V. S. Shmidt, Z. I. Nikolotova, et al., "Physicochemical foundations for
the optimization of the structure of the extractant for regeneration of spent nuclear
fuel," Dokl. Akad. Nauk SSSR, 274, No. 5, 1139-1144 (1984); At. Energ., 58, No. 1,
38-43 (1985).
28. D. Crouse, W. Arnold, and F. Hurst, "Consolidated fuel-reprocessing program alternate
extractants to tributylphosphate for reactor fuel reprocessing," in: Proc. ISEC-83,
Denver (1983), pp. 90-91.
29. E. Horwitz, H. Diamond, D. Kalina, et al., "Octyl(phenyl)-N, N-diisobutylcarbamoyl-
methylphospfine oxide as an extractant for actinides from nitric acid waste," ibid.,
pp. 451-452.
30. W. Schulz and L. McIsaac, "Removal of actinides from nuclear fuel reprocessing waste
solutions with bidentate organophosphorus extractants," Transplutonium-1975, North-
Holland, Amsterdam (1976), pp. 433-477; Proc. ISEC-77, Vol. 2, Toronto, pp. 619-629.
31. E. Horwitz, A. Muscatello, D. Kalina, et al., "The extraction of selected trans-
plutonium (III) and lanthanide (III)- ions by dihexyl-N, N-diethylcarbamoylmethylphos-
phonate from aqueous nitrate media," Separ. Sci. Technol., 16, No. 4, 417-437 (1981).
32. D. Kalina, E. Horwitz, L. Kaplan, et al., "The extraction of Am (III) and Fe (III)
by selected dihexyl-N, N-dialnylcarbamoylmethyl - phosphonates-phosphinates - and
phosphine oxides from nitrate media," ibid., No. 9, 1127-1145.
33. T. Ya. Medved', M. K. Chmutova, N. P. Nesterova, et al., "Dialkyl (diaryl)[dialkyl-
carbamoylmethyl]phosphonic oxides," Izv. Akad. Nauk SSSR, Ser. Khim., No. 9, 2121-2127
(1981).
34. M. K. Chmutova, N. P. Neserova, N. E. Kochetkova, et al., "Extraction and concentration
of transplutonium elements from nitrate media by diphenyl[dialkylcarbamoylmethyl]phos-
phinic oxides," Radiokhimiya, 24, No. 1, 31-37 (1982).
35. A. M. Rozen, "Thermodynamics of extraction equilibria of uranyl nitrate," At. Energ.,
2, No. 5, 445-458 (1957); in: Extraction (in Russian], No. 1, Atomizdat, Moscow pp.
6-87.
36. A. M. Rozen, Z. I. Nikolotova, N. A. Kartasheva, et al., "Complexification of americium,
curium, and lanthanides with organic dioxides and the problem of anomalous aryl
strengthening of the complexes," Radiokhimiya, 19, No. 5, 709-719 (1977).
37. A. M. Rozen, Z. I. Nikolotova, and N. A. Kartasheva, "Method for extracting and con-
centrating actinides and lanthanides," Inventor's Certificate No. 601971, Byull.
Izobr., No. 35, 2571 (1979).
38. A. V. Bogatskii, N. G. Luk'yanenko, V. A. Shapkin, et al., "Extraction of picrates
of alkali and alkaline-earch metals with macroc.yclical esters," Zh. Org. Khim., 16,
No. 10, 2057-2059 (1980).
39. V. V. Yakshin, E. A. Filippov, V. A. Belov, et al., "Coronas in extraction of
uranium and actinides from nitrate solutions," Dokl. Akad. Nauk SSSR, 241, No. 1,
159-162 (1978).
40. V. V. Yakshin, V. M. Abaskin, and B. N. Laskorin, "Electron-donor properties of
macrocyclical polyesters," Dokl. Akad. Nauk SSSR, 244, No. 1, 157 (1979).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
MATHEMATICAL MODEL OF THE TEMPERATURE FIELD AROUND A BOREHOLE
WITH RADIOACTIVE WASTES AND ITS EXPERIMENTAL VERIFICATION
IN FIELD CONDITTONS
E.
G.
Drozhko,
V.
I.
Karpov,
A.
S.
Stepanov,
I.
I.
Kryukov,
V.
F.
Savel'ev,
V.
V.
Kulichenko,
V.
A.
Bel'tyukov, and A. A. Konstantinovich
Radioactive wastes must be reliably isolated from the environment. At present, the
most widespread method is storage in geological formations [1-4]. In the burial of wastes
with a high level of energy liberation, the reliability of their isolation is largely de-
termined by the thermal effects on the rock of the geological mass. As a result of these
effects, temperature stress arises in the rock and may exceed the limiting permissible value
under certain conditions. This may lead ultimately to loss of structural integrity of the
store. In connection with this, the examination of burial options demands careful analysis
of nonsteady temperature fields in the rock mass around the waste site. For most versions
of burial, this analysis is only possible by means of finite-difference calculation schemes,
for various reasons. In waste burial in mine galleries, the use of afinite-difference
scheme is due to the dense lattice of boreholes with wastes, the dependence of the thermo-
physical parameters of the rock on the temperature, and the disrupted structure of the rock
mass at the time of mine construction. However, with a simple burial scheme (for example,
storage of the wastes in a system of deep boreholes [5]), the accuracy of analytical solu-
tions of the heat-conduction equations may be sufficient for practical purposes. The use
of deep boreholes allows high values of thermal load per unit area of the field and efficiency
of the mine workings to be attained. The boreholes may be sunk at a distance excluding
their mutual thermal influence, at least in the period of formation of maximum rock tempera-
ture.
The temperature field in a rock mass from a single borehole with radioactive waste
may be estimated using the model of a linear source. The equation for a source of limited
size perpendicular to an isothermal surface of a semiinfinite mass with thermal power varying
according to the law of radioactive decay may be obtained by the instantaneous-source method
rp
P ~~ i ~ ( ~)
0
~ (z, ti) = erf (2 Doti -}- erf (2 }~lnif -
t -erf (Z ~ bra-ti )f~., -erf. (z 2 7~nT )~4 ~
where t(r, z, T) is the rock temperature at the point with coordinates (r, z) at time T;
to, surface temperature of the rock; k, a, thermal conductivity and thermal diffusivity
of the rock; Q, initial thermal power of the source; ~, radioactive decay constant; L, length
of the heat source; Z, depth of the source.
Using the relation [6]
l .a erf i d, - 2 arsh
I ? (R lea) ~ -= R~
Eq. (1) for a source of constant power (.l = 0) may be written in the form
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 422-425, December, 1985. Orig-
inal article submitted October 30, 1984; revision submitted April 24, 1985.
994 0038-531X/85/5906-0994$09.50 p 1986 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
Fig. 1. Distribution of boreholes 1-7 (a)
and of temperature sensors in boreholes 1
(b) and 2, 4-7 (c): the filled squares
correspond to thermocouples and the. filled
circles to resistance thermometers.
(IS'
t (r' z' ti) - T (r' a) - 8nl,k ~ l i' '~ (z, 't') di'
T (r, z) = 4 Q,~ ( arch ~~ + arsh 2r -
-arch ~r -arsh ~r ) ,
where Ri are the coordinate indices of the given point.
Using the function T(r, z), the steady temperature field around a linear heat source
is described. It has previously been used to estimate the temperature conditions of radio-
active-waste burial (7]. At a great thermal--source length and considerable depth the equa-
tion for a nonsteady problem :must be used. -
The possibility of using Eq. (1) to.estimate the temperature conditions of waste burial
in a system of deep boreholes has been experimentally investigated in field conditions.
For experimental verification, seven boreholes were sunk in loam (Fig. 1). In boreholes
1 and 3, electric heaters of diameter 85 nm and length 1.5 m connected to an ac grid of
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
d t; C
72
d t,,?C
20
96 Z, days
96 2;days
Fig. 2. Dependence of the temperature dif-
ference on the time for upper (a), middle
(b), and lower (c) points of the resistance-
thermometer battery in the boreholes: con-
tinuous curves correspond to calculation
and points to experiment.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
,~~
Fig. 3 Fig. 4
Fig. 3. Surface temperature of borehole (1)
and heater (2); the curve corresponds to
calculation and ttie points to experiment.
Fig. 4. Surface temeprature of heater with
brief disconnection of the .electrical supply
(T < 20 h): curves correspond to calculation
and .points to experiment.
potential 220 V were installed. After installing a bank of resistance thermometers, the
boreholes were filled with the-earth removed when they were dug. The thermometer readings
were recorded by secondary instruments.
The basic aim of the work is to determine the influence of the heating-element dimen-
sions and its depth on the conditions of nonsteady-field formation in the loam, in conditions
of constant heater power, which is the simplest to investigate in field conditions. In
addition, with a more complex dependence of the heat liberation, additional errors appear,
associated with the need to maintain it accurately over time.
In the first stage, a heater of 400 W in borehole 3 was switched on, and operated for
14 days. The readings of the resistance thermometers on switching off the heaters were
measured for a further 17 days after this, which offered the possibility of making mea-
surements during the stepwise change in heater power. Subsequently, the heater in borehole
1 was switched on, and ran for more than three months until a steady temperature distribution
was established in the soil.
The resistance-thermometer readings are analyzed using Eq. (3). To eliminate the in-
fluence of daily and seasonal variations in soil temperature, the difference in the readings
of sensors positioned at the same level in any two boreholes are considered. The thermal
conductivity and thermal diffusivity of the soil in natural conditions before the measurements
was not determined. To estimate these parameters, the readings of the resistance thermometers
in boreholes 2 and 6 at the level of the central cross section of the heating element are
used. The thermal conductivity is estimated from the expression for a steady distribution
[7], which gives 1.71 ? 0.27 W/m?K [8]. The thermal diffusivity of the soil is determined
on the basis of Eq. (2) for a nonsteady temperature distribution in the soil. A theoretical
dependence of the dimensionless temperature difference 026 on FO2 = aT/r2 is obtained here
SnLk
Ozs = [tz (rz, zo, '~) - is (rs, Zo, ti) ] P ( 4 )
where ti(ri, zo, T) is the soil temperature at the point (ri, z); i is the borehole number.
On the basis of the resistance-temperature readings and in accordance with the pre-
liminarily determined thermal conductivity, 026 is calculated, and then from the graphical
dependence FpZ is determined for a fixed time. According to the estimate, the thermal dif-
fusivity is (0.46 ? 0.09)?10-6 m2/sec. The differences in readings of the other resistance
thermometers at the same level are calculated for the resulting values of the soil thermal
conductivity and thermal diffusivity (Fig. 2).
Analysis of the results of surface-temperature measurements for the heating elements
shows that when aT/ro ? 1, where ro is the borehole radius, Eq. (3) may be used to estimate
the surface temperature of the borehole. At the same time, the temperature difference be-
tween the borehole and heater surfaces may be determined on the basis of the relations of
steady heat transfer (Fig. 3). The wall temperature of the borehole is calculated from
(1) and the temperature difference in the air gap between the heater and the borehole wall
from the relation for the radiant heat transfer for the given emissivity of 0.7.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
Initially, when (z - Z)/(Z/aT) > 3, the surface temperature of the heater may be calcu-
lated from the model of an electric cable of infinite length [9]. In Fig. 4, the results
of measuring the readings of the central thermocouple of heater 1 on brief disconnection
are shown together with the results of calculation on the basis of the electric cable model.
Field measurements show that the given model of a linear source is completely satis-
factory in describing the formation of a nonsteady temperature field around a cylindrical
heat source sunk vertically into a mass beginning at times satisfying the condition aT/r2 ?
1. With radioactive-waste burial in a deep borehole, this condition is satisfied for 1-2
months after loading. The energy liberation in the waste determined before burial by the
radionuclides 90Sr and 137Cs is practically unchanged in this period. The initial period
in which the linear-source model is inapplicable for temperature-field calculation in a
rock mass is short and does not influence the conditions of maximum temperature formation
in the rock determining its limiting thermostress state. If the thermal load on the rock
is chosen so that the temperature variations in the thermophysical parameters of the rock
are within the limits of accuracy of their determination for the rock as a whole, in this
case the model of the linear source may be used to analyze the schemes of waste burial with
sufficient accuracy for practical purposes, for example, in waste burial in rock at thermal
loads of up to 300=400 W/m of borehole length.
1. A. S. Nikiforov, V. V. Kulichenko, and M. I. Zhikharev, Safe Storage~of Liquid Waste
from Atomic Power Plants and Radiochemical Production [in Russian], Energoatomizdat,
Moscow (1984).
2. t. I. Kryukov, V. V. Kulichenko, and Yu. P. Martynov, "Conditions of burial of high-
activity solidified waste," in: Research in Spent-Fuel Reprocessing [in Russian],
Vol. 2, COMECON, Prague (1972), p. 34.
3. N. N. Verigin, Yu. P. Marynov, and I. I. Kryukov, "Nonsteady temperature fields in
soil burial of high-activity solid wastes," in: Research into the Safe Storage of
Liquid, Solid, and Gaseous Radioactive Wastes and Deactivation of Contaminated Sur-
faces [in Russian), Atomizdat, Moscow (1978), pp. 129-131.
4. V. V. Kulichko, N. V. Krylova, and I. I. Kryukov, "Properties of highly active wastes
determining their behavior on burial in geological formations," in: Underground Dis-
posal of Radioactive Waste, IAEA, Vienna (1980), pp. 201-207.
5. T. Ringwood, "Safety and depth for nucler disposal," New Sci., 88, No. 1229 (1980).
6. N. N. Verigin, et al., Hydrodynamic and Physicochemical Properties of Rock [in Russian],
Nedra, Moscow (1980), p. 66.
7. N. N. Verigin, V. V. Kulichenko, Yu. P. Martynov, and I. I. Kryukov, "Self-heating in
the underground burial of solid radioactive wastes," in: Underground"Disposal of Radio-
active Waste, IAEA,.Vienna (1967).
8. A. F. Chudnovskii, Thermophysical Characteristics of Disperse Materials [in Russian],
Gos. Izd. Fiz.-Mat. Lit., Moscow (1962); pp. 430, 431, 434.
9. H. S. Carslow and J. C. Jaeger, Conduction of Heat in Solids, Oxford University Press,
New York (1959).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
PASSAGE OF PRIMARY PROTONS THROUGH A SHIELD WITH A RANDOM
DISTRIBUTION OF THE MATERIAL
V. G. Mitrikas, V. M. Sakharov, UDC 539.125.42
and V. G. Semenov
When the radiation conditions inside spacecraft are calculated, one usually employs
the assumption that the spacecraft can be modeled by a few sectors with a constant thickness
of the material in each of the sectors [1] and that the proton dose is the superposition
of the-doses developing behind infinite plane layers of a homogeneous material the thickness
of which is equal to the thickness of the material in the sectors.
The basis of this approach to the calculation of the absorbed proton dose inside a
spacecraft has been considered in a large number of papers [1]. But there exists practically
no information on the influence of the inliomogeneities of the material due to the design
and the equipment of the spacecraft on the development of the dose. We consider in the
present work the results of investigations in which the development of the doses in the
equipment was studied, with the equipment being characterized by a randomly inhomogeneous
distribution of the material. The investigations were made for the case of normal incidence
of monoenergetic protons on the equipment.
The calculation of the primary proton spectra in the bulk of the equipment is-based
on the assumption that the equipment can be described by a random function in analytic form.
The types of functions have been theoretically explained in [2]. We consider for the sake
of simplicity functions of the Rayleigh type in which the probability of the material thick-
ness x (g/cmz) is represented for the geometric dimensions z (cm) in the following form:
~lz L 2~1z J
where yz denotes the average value of the thickness; qz denotes the dispersion of the dis-
tribution; and the subscript z indicates a parametric dependence. With the range-energy.
relation for protons, R =.a In (1 + bEa), where a, b, and a denote the constants depending
upon the material of the equipment [3]; the material thickness x through which the radiation
passes can be expressed as
x = a In [(1-i- bEo')/(1 -{- bE?G)),
where E~ and E denote the energy of the protons incident upon the layer and behind the layer
of thickness x, respectively. Since x is a random quantity which can be characterized by
the distribution function of Eq. (1), the density of the proton distribution over the energy
is
Yz (~') =~ (E+ ~o)-Yz ex f - ~~ (E, Eo)-Yz)z l abaEa-1
~i p l 2~1i J 1-~- bEa
behind the layer z.
When we derived Eq. (2), we used the well-known form of expressing the probability
density of the distribution of a random quantity through another random quantity related
to the first one by a dependence in the form of a function.
In real blocks of equipment, the range of possible thickness values on which the
probability density is defined has its lower limit given by the xmin value (e.g., the sum of
the thicknesses of the mounting plates) and its upper limit given by xmax? The normaliza-
tion coefficient for Pz(E) has the form
A _1 _.eXp (' _ (Yz-zz m)2 J - exp [ - (~'m;~X 2 Y~)2
I 2~nz .J 2nz
(3)
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 425-428, December, 1985. Orig-
inal article submitted April 26, 1984.
0038-531X/85/5906-0999$09.50 ? 1986 Plenum Publishing Corpora*_ion 999
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved FF-or Release 2013/02/20: C~LIA-7~RDP10-021968000300070006-9
..5001 (~ ~ 301~~ a
f0~ ~ ,~ ~ 0
0 70 20 z, cm
Fig. 1 Fig. 2
Fig. 1. Specific absorbed proton dose D in a
solid medium and in blocks of equipment with a
thickness distribution in dependence upon the
depth in the case of a Rayleigh distribution
law: 1) solid medium (p = 0.53.103 kg/m3);
2) block 2; 3) block l; ) calculation.
Fig. 2. ,Activation integral Ai of the threshold
detectors over the depth of block a) 1 and b) 2
of the equipment:
1) 19F (p, pn)
18F;
2) Z'A1
(p, .3n) 24Na; O, ?
tion.
) experiment;
)
calcula-
Equation (2) accounts for the proton losses only by ionization. The approach suggested
by the authors of [4] was used to take into account the attenuation of the proton flux by
nuclear interactions. The probability of a proton passing without nuclear interactions
over a path on which its energy changes from Ea to E as a consequence of ionization losses
is given by the equation
E?
'' ? (E')
k = exp [ - ` dE'/dx dE'
E
or, according to [4],
n
~_ i
In the general case, the proton spectrum in the depth z of a block of equipment is
given by the product of Eqs. (2), (3), and (4):
dN x (E, Ea (~ (F,'-E~)--yi] ?6aEa-1 n
z, E) _ )-YZ ex { } A ~' C ex ,c A E )
dE ( o ~1= P 2rli 1-)-bE" : P [-!~ eff ?' ( ~+ o ~?
4-1
Having obtained in this fashion the proton spectrum, we can more readily calculate
the required functional relationship either in the form of the absorbed dose
D (z, Eo)-1.6.10-" ~ dN dB dE,
aE a:~
where dx denotes the ionization losses, or in the form of the activation integral
(4)
At (z, Eo) _ ~ dE 0: (E) dE, ( 6 )
where 0i(E) denotes the cross section for the activation of the i-th element by protons.
Figure 1 depicts the results of a calculation of the depth distribution of the absorbed.
dose of protons with Eo = 100 MeV in solid aluminum and in randomly inhomogeneous media
for distribution functions Pz(x) of the type of Eq. (1) with the parameters y(z) = pz -
1.910z, r~2(z) = 2.2702; the values of the parameters were determined from the results of
gamma thickness measurements on blocks of equipment [2].
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
N
1~
V
C
O ~
i -`tea i i i i
70 75 0 70 20 30 x
cm
,
x, g/cm2
Fig. 3 Fig. 4
Fig. 3. Functions f(x) of the probability density
of the thickness distribution in block 2: 1)
Rayleigh distribution; 2) normal distribution;
and 3) binomial distribution.
Fig. 4. Depth distribution of the specific ab-
sorbed dose in .block 2: 1) normal distribution;
2) binomial distribution; 3) Rayleigh distribution;
calculation according to-the analytic law
of the thickness distribution; s) calculation,
according to the thickness distribution indicated
in Table 1.
The calculation for the solid medium was made taking into account the ionization losses,
the losses of protons by nuclear interactions, range straggling, and multiple Coulomb
scattering. The data of Fig. 1 clearly illustrate the basic difference in the development
of the depth distribution of the dose inside solid media and randomly inhomogeneous media.
It follows from the analysis of these results that in the case of a randomly inhomogeneous
medium, one can ignore range straggling and multiple Coulomb scattering vis a vis the disper-
sion of the distribution of the material in the equipment when the depth distribution of
the absorbed dose is calculated.
The proposed method of calculating proton spectra and the corresponding functional
relationships in blocks of equipment were verified in the Institute of High-Energy Physics
on the linear I-T00 proton accelerator (Eo = 100 ? 0.5 MeV). The calculated values of the
corresponding activation integrals were obtained with Eqs. (5) and (6). The blocks of equip-
ment were in the experiment scanned in the proton beam for modeling a plane multidirectional
source. Activation detectors made from 100-?m-thick LiF and 91 were mounted in the depth
of a block along the proton beam path. Ten detectors were mounted at each fixed depth of
the block on a plane perpendicular to the beam. After the exposure, the activity of the
products from the reactions 3Li(p, n)~.Be (threshold tit MeV), 19F(p, pn)98F (threshold ti10
MeV), and 13A1(p, 3pn)iiNa (threshold ti30 MeV) was measured.
The confidence interval shown in Fig. 2 corresponds to the dispersion of the distribu-
tion function of the thickness of the material at a fixed depth in the blocks and reflects
the possible error of the functional relationship under inspection at a specific point inside
a block. This interval is much greater than the error made in the determination of the
activity of an individual detector and also exceeds the average values of the activity in
the particular depth.
It is interesting to determine firstly the influence which the analytic representation
of the thickness distribution of the material has upon the proton dose (distribution used
to smooth the results of the gamma thickness measurements) and secondly the requirements
to the.accuracy of determining the parameters of the analytic representation and their de-
pendence upon the geometric dimensions; it is also of interest to determine the errors which
arise in the calculation of the proton dose and which are associated with the replacement
of all materials composing the equiment by an aluminum equivalent.
The test calculations were made for the incidence of a beam of protons with the energy
E~ = 100 MeV on a typical block with the average density 5.102 kg/m3. The experimental
thickness distributions of the material in the block were smoothed by normal or binomial
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
990
~ ~ \ ~ 190
~ 70 ~ ~ \ \\ q ~ 920
a 60
i ~ - i
D 90 20
Fig. 5
i ~ i
30 Z, cm
10 20
Fig. 6
Fig. 5. Depth distribution of the specific absorbed dose for various
values of the parameters of the distribution function: Y = czz
(O ^ x 0 ? refers to a = 0.25, 0.293, 0.325, 0.36, and 0.39, re-
spectively; b = 0.139); ---) 02 = bz (b = 0.0815, 0.139, 0.163, or
0.202 (the curves merge within the error limits indicated); a =
0.325).
Fig. 6. Depth distribution of the specific absorbed dose for
various materials of a block with the density p = 0.53.103 kg/m3;
1) tissue; 2) A1; 3) composition of materials with Weff = 17.7; 4)
Fe.
TABLE 1. Parameters of the Distribution
Laws of the Material Thickness in a
Typical Equipment Block
Normal
Rayleigh I
Binomial
distribution
distribution
distribution
f (~, z) _
t (~, z) _
f ~~, z) _
= 1 X
6Z ~27L
X es
pC-~x-xz~2_~;
26z
a = pz; az =1,844 x;
xmtn=0,1 z/1,9;
zmax=z-f-~
_(x-Yz X
~i
XCX ~ -~~-~z~"~.
P Ana
1'z-.z-1,91 6z ;
tt=a/11; N=z/Il;
distributions and also by the Rayleigh distribution (see Table 1 and Fig. 3). The Rayleigh
law renders the most satisfactory distribuiton.
Figure 4 illustrates the dependencies of the specific absorbed dose of primary protons
in the depth of the equipment block in which the distribution of the material is described
by the distribution functions listed in Table 1. The binomial distribution describes with
great errors "tails" of the distributions; the maximum of the depth dependence of the dose
differs from the maxima of the other distributions by about 207 and is shifted toward smaller
depths in the block. In the case of the normal distribution, the proton dose attenuation
resembles the Rayleigh distribution though the normal distribution only inaccurately describes
the experimental distribution of the thickness in the range of small values. This range
is characterized by a slight spread of the initial proton energy and by irrelevant attenua-
tion by nuclear interactions. Accordingly, the depth dependencies of the dose are practi-
cally the same at low thicknesses of the material for these distribution functions. The
proton doses differ substantially at great depths in the block. Thus, for z = 40 cm, D =
9.8 nrad/(proton?cm-2) in the case of the normal distribution (1 rad = 0.01 Gr), whereas
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
in the case or r.ne xayleign aisLriouLion, L = ~.o nraa/lproLOn?cm -~. 1L roltows rrom
this analysis that for a particular energy the thickness range which in regard to the de-
formation of the primary proton energy is large and which has a high stopping power is most
important in the representation of experimental distribution laws by analytic functions.
This conclusion can be qualitatively drawn from an analysis of Eq. (4). Different
changes in proton energy correspond to the same change in x with respect to yz. For x <
yz, the change in the proton energy is small and the dE/dx changes are accordingly small.
For x > Yz the transition to large dE/dx values takes place and, hence, the thickness range
>Yz has a greater influence upon the dose.
The analytic smoothing of the experimental distribution function of the thickness of
the material in the equipment implies that the first initial and the second central moments
of the anlaytic and experimental distributions coincide. It is therefore interesting to
assess the influence of the errors made in the determination of the moments of the function
upon the depth dependence of the specific absorbed proton dose. For the purpose of estab-
lishing-this influence, we calculated the depth distribution of the dose in the equipment
block in which the distribution function of the thickness was. given by the Rayleigh function
of Eq. (1). The average thickness value Yz was varied by ?20~ and the dispersion in the
range ?50~ with Yz = 0.2z being constant. The results of the calculations have shown (Fig.
5) that a 57 error of the 7z value implies a dose error of 10-12~ over the depth of a block.
Similarly, .a 50~ error of the dispersion implies a dose error of 17-20~. Disregarding the
secondary nucleons leads to an error in the calculation of the proton dose and we must there-
fore conclude that the error of the average thickness value of the material must be ~ 5-10~
and the dispersion of the distribution must be < 20-30~.
Aluminum with the atomic number 13 was used in the preceding calculations as the ma-
terial of the block. In order to determine the validity of using aluminum equivalents in
calculations, we studied the influence of the isotope composition in the equipment block
upon the depth dependence of the attenuation of the absorbed dose of primary protons. The
following materials were considered: biological tissue (effective atomic number Weff =
3.4); A1(Weff = 13);Fe.(Weff = 26); and a block consisting of a mixture of C, F, Si, A1,
Cu, and Fe (Weff = 17.7). All calculations were made with the distribution function of Eq.
(1) for the proton energies 100, 72, and 30 MeV. Practically the same depth distributions
of the proton dose were obtained for all the versions with the exception of the biological
tissue (see Fig. 6); a noticeable change in the case of biological tissue is observed be-
cause a large amount of hydrogen is present in the composition of the block. However, taking
into account that a noticeable hydrogen concentration, as in tissue, must not be expected
in the composition of equipment blocks mounted in spacecraft, we may conclude that the use
of the aluminum equivalent in calculations of the absorbed proton dose must not lead to
important errors.
LITERATURE CITED
1. J. Haffner, Nuclear Radiation and Shielding.in Outer Space [Russian translation],
Atomizdat, Moscow (1971).
2. V. V. Bodin et al., "Experimental results of the disribution of the material thick-
nesses in the equipment of spacecraft," in: Reports of the Second All-Union Scientific
Conference on Shielding Installations of Nuclear Technology from Ionizing Radiation
[in Russian], Moscow Inst. of Physics Research, Moscow (1978), pp. 103-104.
3. R. Alsmiller et al., Shielding of Manned Space Vehicles, Report ORNR-RSIC-35 (1972),
p. 99.
4. A. V. Kolomenskii, V. G. Mitrikas, V. A. Sakovich, and V. M. Sakharov, "The effective
attenuation coefficient of radiation in an inhomogeneous medium," At..~nerg., 44,
No. 6, 517 (1978).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
MEASUREMENT OF THE NEUTRON-INDUCED FISSION CROSS SECTION RATIOS
OF 236U AND s3sU FOR ENERGIES OF 4-11 MeV
A.
A.
Goverdovskii,
A.
K. Gordyushin,
B.
D:
Kuz'minov, A.
I.
Sergachev,
V.
F.
Mitrofanov, S.
M.
Solov'ev,
The buildup of 236U nuclei in a reactor core and subsequent radiative capture lead
to the formation of 237Np and 238Pu, which are largely responsible for the activity of spent
fuel, and to the breeding of s3sU nuclei in the (n, 2n) reaction. Therefore, a strict
requirement of a 27 error is imposed on the most important characteristic of the isotopic
balance - the fast-neutron fission cross section [1]. Since this requirement is not met
by the presently available experimental values of of, particularly for En > 4 MeV, they
need to be extended.
The 236U and z3sU fission cross section ratios were measured for neutral energies of
4.24-10.69 MeV by pulse synchronization [2] at the EGP-lOM FEI accelerator. The neutron
source was the D (d, n) 3He reaction in a gaseous deuterium target at a pressure of (1-1.2)?
lOs Pa. The energy range 4.24-5.6 MeV was spanned by varying the angle between the directions
of motion of the deuterons and neutrons.' For a total resolving .time of 3-4 nsec (full width
at half-maximum) the main and background events were separated by time of flight with a
0.7 m flight path: The fission fragment detector was a back-to-back ionization chamber
filled with a mixture of argon and methane to a pressure of 1.4?lOs Pa. The chamber plates
were 2 mm apart, and the field intensity was 2.5 kV/cm. The chamber housing was made of
silver plated brass whose linear dimensions were calculated by Monte Carlo methods to mini-
mize neutron scattering from structural materials. The targets containing the fissile iso-
topes were fastened to one another in the chamber by backings so that at a distance of 50-60
cm from it the difference of the neutron flux at the nearest and farthest targets did not
exceed 0.2-0.37. The targets were prepared by depositing uranium oxides from aqueous solu-
tions on thin aluminum backings which were subsequently annealed. Their isotopic composition
was determined by a mass-spectrometric method (Table 1), and the nonuniformity in thickness
(107) was measured with a miniature silicon semiconductor alpha detector by scanning along
a radius of the active spot.
We found the ratio of the numbers of 236U and zssU nuclei in samples No. 1 and No.
2 respectively by normalizing the energy dependence of ofe/ofs by the method of isotopic
admixtures: pairs of targets 4/3, 5/3, and 6/3 were irradiated in turn in a flux of fast
neutrons with energies of 7.34, 8.10, and 8.91 MeV, and by neutrons slowed down to 0.5-0.6
MeV by a layer of polyethylene 20 cm thick. The region of normalization was determined
on the stable "plateau" as a function of of6/ofs above 7 MeV. Corrections were introduced
into the results of the absolute measurements to take account of factors which distort them:
the complete stopping of part of the fragments in the target (dl), the background of
scattered neutrons (82), the difference of the efficiencies of recording fragments by the
chambers with z3sU and z36U (S3), the nonuniformity of the neutron flux (S4), the uncorrelated
neutron background in the laboratory and the instability of the electronic recording circuit
(Ss), the fission of impurity isotopes (86). Only the energy dependent corrections S1,
d2, Ss, and d6 were introduced into the unnormalized values of the fission cross section
ratios obtained with targets No. 1 and No. 2 for the same En. The ratios of the values of
of6/ofs obtained for pure targets and targets with an isotopic mixture gave the following
values of the normalization factor KN: for energies of 7.34, 8.10, and 8.91 MeV respectively,
1.142 ? 0.008, 1.157 ? 0.008, and 1.138 ? 0.008.
The procedure for introducing corrections was discussed in detail in [3], and the use
of the method of isotopic admixtures in [4]. Typical values of the corrections and the
errors they introduce into the measured values are the following:
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 429-432, December, 1985. Orig-
inal article submitted February 1, 1985.
? 1986 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
TABLE 1. Isotopic Composition of Targets Used
correction
value
error
S1
0.2-0.3
-
S~
(1.25
-
S3
0.3-0.4
O.li-0.8
S9
1.0-1.7
0.2
SS
1.0
0.95
S~
0.95
-
The correction S1 is small in spite of the appreciable thickness of the targets (0.2-
0.5 mg/cm2). This is related to the use of the method of .isotopic admixtures for determining
the absolute energy dependence of Qfe/ofs. In other cases under these same conditions S1
can reach several percent.
The possible inhomogeneity of the isotopic mixture can lead to an uneven average depth
of deposition of the zssU and zseU nuclei, which affects S1. The inhomogeneity was deter-
mined by a method similar to that used earlier in [5]. The result was negative (the mixtures
were homogeneous).
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
,~-,
o~
o~
o m ob? o 0
w
o? ~ oE: ~ 1 ?a ? o
u o~? }pbo oodiv "u~~~ ~
~$~ o 0 0
0 0
~
o ~
u
~~ ~~
~
o
~?pt 9?O~
?t
o _
Fig. 1. Measured values of neutron-in-
duced fission cross section ratios of
sseU and z3sU: ?) our work; O ) [6];
In processing the results of the measurements particular attention was paid to the
separation of the sources of systematic and random errors and their estimate. The statisti-
cal error was estimated from the spread of values of the ratios of the 236U and zssU frag-
ment counting rates, taking account of corrections at various levels of discrimination in
the corresponding recording channels. It amounted to 0.6-1.2~. The error of the fragment
counting efficiency DE was found by extrapolating the pulse-height spectrum to a zero level
of discrimination. It reached 0.5-0.8~ depending on the filling of the ionization chamber
and the thickness of the sample. The error ds is statistical, since it is determined by
the statistics of the spectrum set in the transition part between fragments and alpha parti-
cles. At the same time the extrapolation is based on a model approach whose uncertainty
is difficult to estimate, and hence a decrease of the error 0E does not follow from the
observed constancy of the efficiency over the whole range of En. For this reason the ratio
of the counting efficiencies and its error were determined separately for each point, and
DE entered the total error of the measurements quadratically. The situation is similar for
the error Ds in measurements on moderated neutrons. The error of the normalization of the
energy dependence of 6f6/ofs at reference points was 0.51, and was determined from the
spread of the normalization factor.
The error of the separation of spurious events related to all the components of the
neutron background, including the part of them determined in measurements with an evacuated
target, was 0.1-0.6~, and was random (the statistics of the set of corresponding parts of
time-of-flight spectrum). The remaining errors - corrections, dead time of the recording
channel, the instability of the timing etc. - are negligible (Table 2). It is clear from
Fig. 1 that the spread of the experimental points reaches 6-8~, which is considerably larger
than the listed errors of the mesurements. The character of the spread of the data was
investigated in their statistical analysis, and consisted of two parts:
the determination of the correlations of the energy dependences in the range 4-9 MeV
(the range spanned). A calculation of the corresponding correlation matrix showed a linear
functional dependence of the experimental data, i.e. the collection of points of various
experimenters can be displaced along both the energy axis and the axis of ordinates;
an estimate of the relative displacement of the sets of data along the of6/of6 and
En axes.
For the first case averaging of6/ofs over the energy range 5.5-8.5 MeV gave the follow-
ing values: 1, our work; 1.021, Behrens and Carlson [6]; 0.978, Meadows [7]; 0.974, Konde
[8] (normalized to our data). The displacement on the energy scale was determined by
analogy with [9]. To do this the cross section ratio was converted to the fission cross
section by using the standard data on ofs [10], and the data of each author were described
by a smooth curve in the region of rise to the second plateau
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
4 S 6 7 E,,, MeV
Fig. 2. Results of describing ex-
perimental data by a smooth curve.
.~
s-?--o------~.
~o
.
~( ? ..
~? 00 ..r~O 00 O. .
I ? O o
F
00 ~ O O O O
yo o ? ?. ~ . ~ :. d ? b
o ~?. . . g .. ~ o
?ao?.a v?o.~ o . o .a
8 B'ao o
ode $o o?
0 0 0 ? o
4 5 6 7 B 9 E,,, MeV
Fig. 3. Deviations of experimental
data from ENDF/B V and from our es-
timate.
where of_is the value of the cross section on the first plateau, on~n~ is the cross section
for the inelastic scattering of neutrons, p(E) is the neutron emission spectrum, and P
(En - E) is the probability of fission of a nucleus after the emission of a neutron of energy
E. The parameter Tf, the height of the fission barrier entering P (En - E), characterizes
the energy spread, and is ti0.1 MeV. It was observed that by varying the parameters a curve
of the form of Eq. (1) could describe the whole-set of data, which serves as a basis for
estimating the cross section for the fission of zseU by neutrons of energy >5 MeV. Four
sets of the paremters T (the nuclear temperature) and Tf were averaged with weights inversely
proportional to the square of the rms deviations of the experimental values from the corre-
sponding curves. The solid curve of Fig. 2 shows our estimated curve with parameters T
and Tf varying over a wide interval, and the dashed curve with parameters closest to the
most realistic values [11].
The lower part of Fig. 3 shows the deviations of afs/of5 from our estimate, and the
upper part the deviations from ENDF/B V. The error of the estimate values is naturally
determined by starting from the spread of the parameters of the approximation, which leads
to 2-37.
Our analysis shows that the energy dependence of ofb/of5 is known reliably at the
present time. The absolute values calculated from the cross section ratios have a 5-8~
spread, which obviously is determined by the systematic errors of the various studies and
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
a. aac a. ca:i ir,a a.ivaa aac ri v~.cuua_c vi iov ~.viri... wcisaa a. iaaF, Wll.la 111 a. aac iiaua~cwvl.n vi a.aac auethod
of isotopic admixtures. On the whole it can be concluded that the existing set of data
ensures an error of the estimate of .of for En > 4 MeV at the 57 level (taking acocunt of
the error of the standard of).
The authors thank N. V. Kornilov for valuable discussions of the measurement procedure.
1. WRENDA 83/84. World request list for. nuclear data. Nuclear data section. Vienna:
IAEA (1983).
2. A. A. Goverdovskii, A. K. Gordyushin, B. D. Kuz'minov, et al., "Measurement of the
fission cross sections of heavy nuclei by the method of pulse synchronization," in:
Neutron Physics, Part 4 [in Russian], TsNIIatominform, Moscow (1984), p. 115.
3. A. A. Goverdovsky, A. K. Gordjushin, B. D. Kuzminov, et al., "The z36U and 238U to
z3sU fission cross section ratios in the neutron energy range 5-11 MeV," in: Proc.
Int. Spec. Meeting on Transactinium Isotopes Nuclear Data, Uppsala (1984).
4. A. A. Goverdovskii, A. K. Gordyushin, B. D. Kuz'minov, et al., "Measurement of the
fission cross section ratios of z36U and z3sU by the method of isotopic admixtures,"
Problems of Atomic Science and Engineering, Ser. Nuclear Constnats 3 (57) [in Russian],
(1984), pp. 13-15.
5. A. A. Goverdovskii, A. K. Gordyushin, B. D. Kuz'minov, et al., "Measurement of the
ratio of the neutron-induced fission cross sections of 237Np and z3sU in the energy
range 4-11 MeV," At. ~nerg., 58, 137-139 (1985).
6. J. Behrens and G. Carlson, "Measurements of the neutron-induced cross sections of
234U~ 236U, and 238U to z3sU from 0.1 to 30 MeV," Nucl. Sci. Eng.,. 63, 250-267 (1977).
7. J. Meadows, "Neutron-induced fission cross section ratios of z3oU~ z36U and z3sU,"
Nucl. Sci. Eng., 65, 171 (1978).
8. C. Nordborg et al., "Fission cross section ratios of z3zTh, z36U, and z3sU," in: .Proc.
Int..Conf. on Nuclear Data, Vol. 1, Harwell (1979), p. 910.
9. H. Knitter and C. Budtz-Jorgensen, "Barrier heights of plutonium isotopes from
(n, n'f)," in: Proc. Int. Conf. on Nuclear Data for Science and Technology, Vol. 1,
Antwerpen, Sept. 6-10 (1982), pp. 744-747.
10. ENDF/B V Third Ed. BNL (1979), z3sU (MAT 1395).
11. S. Bj~rnholm and J. Lynn, "The double-humped fission barrier," Rev. Mod. Phys., 52,
725 (1980).
V. S. Zaveryaev, G. I. Britvich,
V. I. Lebedev, V. S. Lukanin,
F. Spurny, I. Potochkova,
and I. Kharvat
Tokamaks at the present time occupy a leading position in research on controlled
thermonuclear fusion with magnetic confinement. The characteristics of the fields of ioniz-
ing radiations are of great interest in the use of such units since the design to be adopted
and the composition of the equipment required for monitoring the irradiation of the personnel
depend upon the fields. Besides that, research on the ionizing radiations can provide use-
ful information for plasma diagnostics.
Our work relates to research on the T-10 unit as a source of radiation and provides
an evaluation of the efficiency of the radiation shielding. The experimental results were
obtained during several operational cycles of the T-10 by the co-workers of the Institute
of High-Energy Physics (1977) and the Institute of Radiation Dosimetry of the Academy of
Sciences of the Czechoslovakian SSR (1981-1983) together with the co-workers of the I. V.
Kurchatov Institute of Atomic Energy.
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 432-436, December, 1985. Orig-
inal article submitted February 1, 1985.
0038-531X/85/5906-1008$09.50 ? 1986 Plenum Publishing Corporation
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
Fig. 1. Overall view of
the T-10 unit: 1-8) points;
X, Y, and Z) directions in
which the detectors were
moved.
Short Description of the T-10 Unit and of the Conditions Studied. The T-10 unit is
a cruciform transformer the secondary "winding" of which consists of deuterium filling
a toroidal vacuum chamber [1, 2]. The large radius of the torus is 150 cm, its small radius
39 cm. The inner chamber is inserted into an external stainless-steel chamber with a radius
of 50 cm. A 5-cm-thick toroidal copper shield is placed between the chambers. The outer
chamber is surrounded by the sturdy coils of the longitudinal magnetic field (Fig. 1). In
order to reduce the interaction of the plasma with the walls, tungsten diaphragms (1977
and 1981) and graphite diaphragms (1982-1983) were employed. The unit is surrounded by
a shadow-type radiation shield of heavy concrete (density 3.6 g/cm3) with a thickness of
1 m and a height of 5 m.
A discharge pulse has a length of 1 sec; the plasma current is 200-400 kA. At a suf-
ficiently high temperature (0.6-0.8 keV) and density ((5-8)?1013 cm-3) of the ions in the
.plasma, the thermonuclear d + d reaction takes place and emission of neutrons with an in~
tensity of 1010 sec-1 is observed during ti0.6 sec. The total neutron yield amounted to
(1-8)?109.. Such discharges are termed thermonuclear discharges.
In certain cases part of the plasma electrons were transferred into the continuous
acceleration mode and reached an energy of several dozen megaelectron-volts. Being incident
on a diaphragm, these electrons generated bremsstrahlung and photoneutron emission_on a
high intensity level. The neutron yield reached (1-10)?1012. .Such discharges are termed
acceleration-type discharges [3]. In 1977 a series of similar discharges was especially
studied.
In the 1981-1983 experiments, usually the radiation characteristics averaged over
several series of discharges were determined: in the 1981 experiments, acceleration-type
discharges often appeared; in the 1982 experiments, acceleration-type discharges were ob-
served only in the second half of the series of measurements; in 1983, the edge of the plasma
touched on secondary steel objects in the second half of the series of measurements and
therefore a considerable number of unstable and acceleration-type discharges developed.
We describe in the present work the dosimetric characteristics of individual discharges
or average values for a series of T-10 discharge pulses. The unit makes it possible to
obtain 30-40 discharges in a shift.
Measurement Methods. The bremsstrahlung was recorded with a Geiger-iVluller counter
having an extremely low relative neutron sensitivity for work in composite neutron-x-ray
fields [4, 5]; various ionization chambers and the.rmoluminescence detectors (TLD) in the
form of 0.85-mm-thick LiF and ~LiF tablets with a diameter of 4.5 mm were used for the same
purpose.
The neutron radiation was measured with: thermal-neutron detectors (In foil, BF3
counters) surrounded by a moderator for recording neutrons with an energy 20 MeV) along the T-10 chamber
in acceleration-type discharges with a
tungsten diaphragm. Points 1%~-8%~ are
situated on the plane of the vacuum
chamber between the coils of the longi-
tudinal field and are opposite to
points 1-8 (see Fig. 1).
The photonuclear reactions which take place at the diaphragm under the influence of
a beam of accelerated electrons produce unstable isotopes with various half-lives. Radio-
activity measurements made immediately after an acceleration-type discharge have shown that
the radioactivity is concentrated in the region of the diaphragm and that the dose rate
of the radiation on the outer surface of the T-10 chamber proper reaches 20 uR/sec 3 min
after a pulse (1 R = 2.58.10-'' C/kg). When the decay of the activity is brought into account,
the personnel near the chamber can receive the maximum admissible daily dose during 20-40
min.
Spatial Distribution of the Radiations Near the T-10 Chamber. AGeiger-Muller counter .
with energy compensation .by lead and tin filters was used to study the spatial distribution
of the bremsstrahlung in thermonuclear discharges [4). The isotropy of the neutron radia-
tion was determined with a 3He counter in a sphere with a diameter of 12.7 cm. The de-
tectors were mounted facing the gaps between the coils of the longitudinal field at a height
of 0.4 m from the plane of the T-10 vacuum chamber at points 1-8 (see Fig. 1). Measurements
have shown that the gamma radiation field is rather homogeneous (average dose 1.6.10-z mGy
per discharge) though a noticeable signal increase is observed in the region of the diaphragm.
A comparison of the signals of the Geiger-Muller counter with and without filters has
shown that the contribution of photons with an energy below 200 keV to the dose does not
lead to a measurement error in excess of ?20~. This is a logical value when the radiation
is shielded with the construction materials of the unit.
There were only a few thermonuclear discharges in the time of these measurements and
our data suffice only to demonstrate qualitatively that the neutron emission in such dis-
charges is homogeneous.
Investigations of the acceleration-type discharges provide more information on the
radiation fields near the unit. By contrast to the thermonuclear discharges, a strong in-
homogeneity of the neutron emission is observed in this case along the T-10 chamber (see
Fig. 2) and sharp peaks appear in the region of the diaphragm (points 1%~ and 5%~). The
.fluctuations are less pronounced for neutrons with En > 20 MeV. This means that the dia- ~
phragms are of great importance for the development of the radiation.
When the tungsten diaphragm was replaced by a graphite diaphragm, the level of the
gamma radiation was reduced and the flux of the fast neutrons (En > 1 MeV) decreased by
a factor of 100 (see Table 1). The rather large raidation dose observed in 1983 can be
explained by a steel rod which was at the edge of the plasma string and acted as a diaphragm.
Nevertheless, the radiation level was still much lower than in the 1981 experiments in which
the tungsten diaphragm was employed.
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
i .r
n.
x
J
w ~p
i
Number of the point
a~
Q) ~
~
10'
o
t!
~
O J
~,
'd ~0
O
Y
M
O
X /
/.
/.
~ ` \
~ ~.~ ~~'
m ~~ 6
0 5 70 75 10
Distance (m) from the center of the unit
Fig. 3 Fig. 4
Fig. 3. Distributions of the neutron flux (En > 1 MeV), of the equivalent dose
(dielectric track detector in contact with 232Th),and of the exposure dose of
the gamma radiation (TLD detectors) along the T-10 chamber; averaging over 402
discharge pulses (1983).
Fig. 4. Equivalent dose at a considerable distance from the T-10 unit in
acceleration-type discharges with a tungsten diaphragm: -t-, C~,.? ) measurements;
X, Y, Z) according to Fig. 1.
TABLE 1. Radiation Levels at Point 7%~ (see Fig. 2). Acceleration-Type and Un-
stable Discharges
Exposure dose tmGr per discharge)
Track density cm-E per discharge)
As above
TLD
DTD with zs2Th
DTD with ~36U
4,5
2,fi
27
4 6.10-a
i,r.~o-~
1,4
2,(1
1,0
3,4
The change in the radiation from point 1* to point 5* (see Fig. 3) is less than in-~
dicated in Fig. 2. This is logical when an additional steel diaphragm is present at point
5%~. Furthermore, there exists a direct correlation between the neutron flux and the ex-
posure dose. The equivalent dose on the surface of the T-10 chamber is smaller than the
exposure dose by a factor of 5-10.
The flux of neutrons with an energy , (2)
whose cross sections >1 b (lb = 10-28 m2) at a:energy of ZHe, ZHe ti 30 MeV. This reaction
cross section enables achieving a sensitivity in the determination of the lead content of
10-"~, but because of the long half-life of Z84Po (138 days) this requires either intense
irradiation over many days or the emitted a particles must be recorded for several months.
In [6, 7] the reaction of the formation of a-active isotopes in the following reac-
tions with heavy ions was investigated:
a cc a
208zPb(18C, 4rz)z88Ra ---> 2R~Rn --~ le1Pu -->
a a a (3)
282Pb(1RC, 6n)28gRa -- zR~Rn --> 18"~Po --~, (4)
The a particle detector consisted of cellulose nitrate, applied to the sample after
irradiation. The energy of a particles from ZS4Po is ti5.3 MeV (1), (2), ti6.2 MeV (3) from
286Rn, and ti6 MeV (4) from ZS$Rn. This is higher than the detection threshold of cellulose
nitrate (ti3 MeV) [3], so that an absorber must be placed between the sample and the detector
in order to reduce the energy of the a particles, which degrades the resolution of the method.
The attainable sensitivity level is equal to ~ 10-v0~, and the attainable resolution is
?'LO}un. These are inadequate for the solution of a number of geological problems. The
sensitivity for the determination of the content and the accuracy of the spatial distribuiton
of lead can be increased by using for the analysis the fission reaction of lead induced
by accelerated heavy ions. The fission cross section is much higher than the formation
cross section of a-active nuclides and for optimally chosen energy and mass of heavy ions
exceeds 2 b [8].
The mass distribution of the fragments from fissioning of lead by oxygen ions, evi-
dently, has a maximum near A = 110 and Z = 45. In addition, a-active products are formed
in the reaction
z82Pb(~g0, 2a) -- LUO h y zAnRa ?--i =86Rn ---~ 2saPo y.
The difficulty of using the fission effect lies in the fact that the fission fragments
must be detected simultaneously with the irradiation of the sample because of the short
lifetime of the compound nuclei formed. This difficulty can be overcome by using the follow-
Translated from Atomnaya nergiya, Vol. 59, No. 6, pp. 437-439, December, 1985. Orig-
inal article submitted July 30, 1984; revision submitted December 17, 1984.
0038-531X/85/5906-1015$09.50
Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
 Declassified and Approved For Release 2013/02/20: CIA-RDP10-021968000300070006-9
Fig. 1. Scheme used for irradiating the
sample: 1) ion beam, 2) track detector;
3) sample; 4) rotating metallic disk;
S) Faraday cylinder with tantalum foil;
6) television camera.
ing irradiation scheme (Fig. 1): a thin (ti10 um) dielectric detector with a detection thres-
hold chosen so that the ions of the primary beam are not detected is applied flush against
the sample. The detector can consist of either natural or artificial mica, which has the
required radiation resistance and a high detection threshold, and it is easy to obtain from
it thin layers with the required area. To eliminate. the characteristic background formed
by the fission-fragment tracks, the mica layers are annealed prior to irradiation j3].
The object of investigation consisted of bituminous dolomites, in which the redistribu-
tion of the carbonate material is accompanied by mobilization of the ore impurity and growth
of gallenite crystals. We did not observe in the chemical study of the- samples in associa-
tion with the lead, gold, mercury, bismuth, and thorium platinoids, whose. fission fragments
for the irradiation conditions used could distort the picture of the lead distribution.
The content of uranium in these rocks is < 10-60~. Thus the contribution of uranium fission
fragments to the radiographic picture is more than two times smaller than the contribution
of lead.
Oxygen ions with an energy of 9.3-MeV per nucleon were chosen for the irradiation.
The mica was 10-15 um thick, which decreased insignificantly (