THE PHYSICS OF DIELECTRICS IN THE USSR

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP80-00809A000600270019-1
• Release Decision: RIPPUB
• Original Classification: C
• Document Page Count: 22
• Document Creation Date: December 22, 2016
• Document Release Date: September 19, 2011
• Sequence Number: 19
• Publication Date: November 30, 1949
• Content Type: REPORT

Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
CLASSIFICAT,ON tat7!f il&IITtIm-
CENTRAI_ INTELLIGENCE AGENCY
INFORMATION FROM
COUNTRY tum
SUBJECT Scientific - Electronics
HOW
PUBL:SHED Monthly periodical
WHERE
PUBLISHED bbscra
DATE
PUBLISHED Oct 1947
LANGUAGE Russian
MIS OOCUNIST CONTAIN/ INTOHIATION ASSlCTINII TMl NATIONAL DIRNSI
OS T U UNITED STATIA WITNN TNI ^UIINI O! ISPIONASI ACT IT
a. O. C.. I I AND AN. AS ..N.M. In at ". NE11" 00
OI 11f CONTACTS IN AM'. NANKIN TO MR, NAUTMOSILI..INI0. 11 PRO.
MINTID IT LAW. I[AODDCIION OI P.R. II NNONNITIO.
DATE DIST. AA'$ Nov 1949
NO. OF PAGES 22
SUPPLEMENT TO
REPORT NO.
THIS IS UNEVALUATED INFORMATION
Jhogineering physics was developed in the 3SR only after the great October
Socialist Revolution, when vigorously growing socialist industry needed new and
Independent roads of development. This is especially true of the--asw'.uranah:or
physics closely coimected with the technique of electrical insulation -- the
physics of dielectrics.
Credit must to given to Soviet physicists for being the first to begin a
cemprehsnsive study of the processes which take place in dielectrics under the
infliwnce of an electrical field. Such a study Is not of scientific interest
alone but is absolutely essential for rational selection and synthesis of elec-
trical insulating arterials.
At present, the results of thb labor of Soviet scientists in the field of
dielectrios mks it possible, first, to analyze in a ntaber of cases comprlen
processes which occur In dielectrics placed In an electrical field; second, to
calculate certain more inpartant dieiectrio constants fran other constants, link-
Inng the basic electrical properties of dielectrice with their molecular structure;
and third, to correct, at fre4uent intervals, he work of chemists and teohnio?lans
In the construction of new high quality dielectrics possessing the adoessary special
properties.
The attainments of Soviet dislectrical pi,ysics significantly facilitated the
developmemt of insulating techniques not only with respect to selection and syn-
thesis of lns?lnting materials, but also in constructing and designing commsrolal
Insulators aAd condensers, and permitted the finding of new Independent methods
for solving practical problem.
Due
FBI
- 1 -
ulb I IKIeuI ION
71
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
In 1922 - 23, work was begun by Soviet scientists in the field of dielectricsa,
a.,a some of their investigations were published in world literature. Certain prob-
lems were clarified concerning dielectric polarization, electrical conductivity and
rupture of gaseous dielectrics, dielectric polarization and dielectric looses of
polarized liquids, and others,, Electrical characteristics of solid dielectrics,
which are most important from the viewpoint of insulation techniques, ware almost
neglected. The individual experimental facts were known; these rel..ted basically
to the so-called "dielectric anomalies" -- a reduction of the current in a solid
dielectric as time passes, with a constant voltage and, in connection with this,
the cp.rnrost noncon*orimity with Ohm's law.
A. Ionic Conductivity of Dielectrics
Soviet physicists, under the leadership of A. F. Iofe who started this work,
dealt, first of all, with the atuiy of conduc';ivity of solid dielectrics. A car?'a
of new experimental facts and rules were established which permitted the explanation
in a general we- of the nature of the conductivity of solid dielectrics and the
causes for the apparent nonconformity with Ohm's law. Later, these works were greatly
expanded and, at present, the nature of dielectric conductivity is significantly
clearer.
The conductivity of the greats- portion of dielsctrice, under conditions wen
below the rupture point, is essentially of an ionic type. This was clearly shown
by a series of investigations devoted to a verification of Faraday's law. For ionic
crystals, glass, lacquered films, etc., it was shown (Pruzhinino) that Faraday's law
is well substantiated. At the same time, the classical method of verifying Faraday's
law, established by Tuband for crystals, was developed and modified for use in cm-
mercial insulation materials (glass, lacquered films, mica). The mechanics of ionic
conductivity of nongaseous dielectrics, due to the work of Ioffe, Frankel' and others?
can now be treated from a single point of view.
Weakly attached ions in dielectrics, (for example, in crystals -- an ionic
lattice, in glass -- ions which are found in loose structural packing, etc.), under-
going thermal agitation can move gradually from one "potential pocket" to another.
Bach ion in such a transfer overcomes a certain potential barrier, depending on the
structure of the dielectric. At the same time, the kinetic energy of the ion is
spent in work of overcoming this barrier -- the work of activation. Even in a liquid
dielectric, similar thermal agitation of the ions can take place. The Ion adheres
to the molecule, enters into its group of atoms and Is thereby attached to it, find-
ing itself in a potential pocket.
However, the ion taking part in the thermal agitation has a certain probability
of breaking away from the molecule. In breaking away, the kinetic energy of the ion
Is used in working against the cohesive force cf the molecule -- the work of activa-
tion. The activation energy, inherent in the ion, is found not only In a solid but
aluo in a liquil body. After moving some distance along a "free path," the ion abain
falls intc, a potential pocket. With the application of an external electrical field,
activated ions are carried in the direction of the field. Tta average velocity of
the ion in the direction of the field, with small fields, is directly proportional to
the field intensity.
Because of this, the conductivity of any nongaseous dielectric can easily be
found with the aid of the expression for mobility with constants which characterize
the molecular structure of the dielectric. It can be expressed in the following way:
_ Sy 2 1
a - ~qX 6kT ve T,
where 'Y equals natural oscillation frequency of the Ion at the place of attachment,
JL equals the number of ion-carriers of current in a cm3, qq equals charge of the
Ion, X equals mobility of the ion, 8 equals mean free path, II equals activation
energy, kr equals energy of thermal agitation. For liquid dieleu`-ripe it is easy
to link the conductivity with the macroscopic characteristic of the liquid-viscosity.
LW
sanitized Copy Approved tor Kelease 2011/0 /1y : C;IA-KUI 0-UUbUVAUUUbUU2IUU1y-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
The viscosity of the liquid is dependent on the motion of the molecules, which can
be treated in the same way as the motion of io..e. While .flowing, i.e., the notion
of liquid layers, the molecules are subjected to the action of certain forces. The
regulated motion of the moleculea under the action of this force also determines
the viscosity of the liquid.
This analogy permits the establishment of a relationship between the conductiv-
ity and the viscosity of the liquid., i.e., it clarifies the long-known experimental
law of Val'den by which the product of the conductivity and viscosity is constant
for a given liquid dielectric and does not depend on the temperature. The dependencm
of ionic conductivity of uwaAseous dielectrics on temperature, in accordance with
the facts net forth, must have the following form:
~r = A e -- P/7
The coefficient A is comparatively independent of the temperature. If various
types of ions can move in a dielectric, or the same type ions but attached in vary
This dependence was experimentally established by &abeko, luvehinakiy, Shiahkin,
Lazarev, and others for a great number of osseous dielectrics; the right side of
the equation consists sometimes of two terse and sometimes of one.
Thus, ionic conductivity of nongaseous dielectrics is sufficiently clear in a
general way.
The question of precisely what kinds of ions move in the dielectric while current
flows, also served as a subject for research by Soviet physicists. It was established
that in liquid nonpolarize4 dielectrics used for the entire insulation (transformer
and vege`eble oil, solvents for insulating lacquers -- benzol, toluene, xylene, etc.),
the basic carriers of the current appear to be ions of admixtures -- contaminated.
In crystalline dielectrics, light and relati~? weakly attached ions of alkali metals
move first. A break in the curve I9 7 = fl T , found for simpler ionic crystals
(of the type NaCI) served for a long while as the subject of a lively discussion.
A school of thought led by Sme)ml took the position that the conductivity of these
crystals can be divided into two parts: defective conductivity which dominates at
low temperatures, and the correct conductivity which takes place only at high tempera-
ture. Defective conductivity is caused by the motion of ions in the defective
places -- distortion of the crystal lattice which must obviously exist in every crystal.
Smekal considered this the decisive factor in defective conductivity.
Soviet physicists, however, (Gokhberg and others) showed experimentally, in work
devoted to explaining the mechanism of crystal conductivity, that the deciding factor
was considered to be defects in the lattice only because the crystals were poorly
cleaned and their conductivity depended an contaminated lone.
Frankel's work in 1926 was the first theoretical investigation of the natural
conductivity of crystals.
At present, it cpin be considered as established that there are two possible basic
explanation of ion transfers in crystals,. namely, (1) transfer of ions between the
units with a further movement in the interlattiee space, and (2) the jump from one
unit to another -- unoccupied unit (movement of "gaps").
The predominance of one type of movement or another is explained on the basis
of appraioable calculations of corresponding differences in energy. For alkaline-
ballde .. s ale it h. tc been es b Ishe that the moat m L*4ble is the neap mnnhwni?
of conductivity.
? Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 The conductivity of crystals at low tempdrature depends to a great degree on
the presence an nature of admixture. Admimmi lone Ere o nervily none weakly
attached than the basic lattice Sons since they ar.~i located, for the a?9et part,
either in inter-lattice space or in defective spots. Only with n gap type of con-
ductivity can a small quantity of admixture ions be located in unoccupied lattice
units and (due to the high degree of attachment) without causing a noticeable in-
croase of conductivity.
It is very difficult to isolate experimentally the notion of ai.?ixed ions
from that of basic lattice ions at the defective place. In connection with this,
the conductivity mechanism presented by Smekal, even if it could take place, is
by no means the only possibility.
A z,od deal was done by Soviet physicists in studying the conductivity of
glace, Shulmrev, Myuller, i?rkin, and others. in pare glass having only one vit-
reous oxide, there are no ions whose movement would allot; conductivity. Actually,
the conductivity of such glass is nearly negligible. Commercial glass has a com-
plex composition. Besides the vitreous oxide, it contains other oxides of alkaline
and alkaline-earth metals whose presence determine the necessary physical-chemical
and technological characteristics of the glass. The conductivity of ccamaercial
glass is primarily determi- d by the presence of alkaline metals.
If, in the composition of the glass, there is introduced an oxide of a single-
valence metal (Na2O, S2o), then, in the formation of the glass, an atom of oxygen
of the oxide combines with an atom of silicon or boron and occupies one of the
apexes of the elementary nucleus. The atom of the alkaline metal, having only one
valence tie, connects with only one atom of oxygen. As a result of this, at the
place where the atom of the single-valence metal is located, a disintegration of
the structure occurs since has no possibility of combining with another atom
of oxygen, maintaining the network. The disintegration of the structure is shown
by the fact that a region of increased potential energy forms around the atom of
the single-valence metal. Thus the introduction of an oxide of a single-valence
metal causes (1) the disintegration of the structure of the glass, and (2) the
presence of weakly attached atoms (or ions) of the single-valence metal. Both
these factors greatly increase the conductivity of glass.
The introduction of bivalent metallic oxides (for example, CaO, BaO, etc.) not
only does not increase but can even decrease the conductivity of the glass. The
atom of bivalent metal, due to the second tie, can combine simultaneously with two
atoms of oxygen. Thus, in the structural network of the glass, disintegration will
not take place. It can a'-so happen that the bivalent metallic atom will join the
ends of the disintegrated network and eliminate the looseness of structure of the
pure glass. A systematic study of the influence of the composition of glass on its
conductivity was conducted by Soviet physicists. As a result of their work, not
only were the above-mentioned viewpoints established but also a series of interest-
ing experimental data was obtained which permitted a rational approach to the selec-
tion of the desired type of glass.
Soviet physicists, even in the first years after the revolution, gave special
attention to the secondary appearances which are related to the passage of a current
through a solid dielectric at low temperatures and which were especially puzzling
for a long time. The matter concerns the reduction, with time, of current pasting
through a dielectric with a constant applied voltage, and the deviation from Okmf'e
law. The effect was discove more than 50 years ago by Pierre Curie. The Motu
of the Soviet physicists managed to clear up its character to a remarkabld degree.
Twenty-five years ago A. F. loffe, in his investigations an quartz, laid the
foundation for the physics of dielectrics,' particularly in studying current decrease
a solid dielectrics and effects related to it. In this investigation he was ohe
first to understand the high-voltage polarization of solid dielectrics and was drawn
to an explanation of the Possibility of accumulation of space charge in a solid body.
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
At present, due to the work of Gokhberg, V. A. Ioffe, 2inel'nikov, Venderovich,
^'y others there is G9 ?ibiltt of Rigt el9 solid die!?Ct^iC aCCO 7,;14 I---
First type--Crystals with a high degree of symmetry; in clear crystals, in
the absence of defects in the crystal lattice, current decrease does not take place
and the distribution of potential is linear.
Fourth type--Amorphous and ceramic dielectrics: the drop in current occurs in
very short time periods (the establishment of dielectric polarization), and also at
high temperatures (molding): the drop in potential is centered basically in the
molded layer.
Fifth type--Nonuniform dielectrics (commercial laminated insulating materials):
the drop of current is related to the accumm*.?ation of chargo on the surfaces of the
nonhcnogeneous material (the classical Maxwell case).
By Ioffe's work it was shown that if the electromotive force of polarization,
Actual deviation from Ohm's law in solid dielectrics begins in very strong
fields.
(The same designations as above).
The density of the current is no longer proportional to the voltage of the
field but instead depends on it according to a complex law; the conductivity in-
creases with an increase in the field voltage, which agrees qualitatively with
Falls known experimental law.
In strong fields (for glass, for example, h2ghcr than 2.106 volts/om), eleo-
trmic conductivity, which is negligible in weak4ields for most dielectrics, id
superimposed on ionic conductivity.
The temperature coefficient of conductivity changes in strong fields. This
follows from the expression for ionic conductivity
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
As is clear from this expressidn, the conductivity, as the roltage of the
field increases, begins to depend lees on the temperatuie.
The question of electronic conductivity of dielectrics in strong fields is
closely connected with the electrical breakdown of dielectrica. See below.
The dependence of electronic conductivity on the voltage of the field was
first exposed in the theoretical work of Ya. I. F4renkel.
B. Dielectric Polarization and Looses
The vast amount of work by Soviet physicists in the area of dielectric polari-
zation and losses can be divided as follows:
The study of abnormal polarization of Rochelle salt and piezoelectricity;
The study of dielectric polarization and losses of polycrystalline dielectrics;
the discovery of a new type of polarization in barium titanate;
The study of dielectric polarization and losses of amorphous dielectrice and
*aigh-molecular weight compounds, which was developed subsequently into an elaborate
investigation dedicated to the study of amorphous compounds;
The study of dielectric losses and polarization of glass and ceramic materials
with the practicel objective of obtaining new high-quality Insulating materials;
these projects are closely tied in with the work of Soviet chemists, technicians
and engineers in the study of dielectric losses of a great number of commercial
insulating materials (oil, lacquers, various compounds, fibrous materials, plastics)
which will not only permit the development of new materials but thn improvement of
known materials as well;
The study of dielectric characteristics of a new group of insulating materials
consisting of silico-organic compounds and the creating of insulating materials with
higher thermostability.
1. The fundamental work of Surchatov, Sobeko, and others, dedicated to a de-
tailed study of Rochelle salts, permitted the eetablickment of a 'series of rules
which, to a large degree, explain the nature of their abnormal polarization. A
group of dielectrics were also uncovered--isomorphous crystals containing Rochelle
salts--having characteristics similar to Rochelle salts in regard to their spon-
taneous polarization and dielectric hysteresis.
These materials were called piezoelectrical.
Kurehatov first presented the theory of' piezoelectric polarization which quali-
tatively explains the established rules. The theories of piezoelectricity, hovever,
are not completely substantiated as yet. Neverthele , the viewpoint of Kurcbatov
and others, In which It is possible to rotate polar molecules in a solid substance,
was very fruitful. It was developed further by clarification of a series of rules
which were found for commercial dielectrics (fibrous materials, oleowax, halowax,
etc.).
2. A good deal of the work carried on in the recert years by Soviet scientists
(Bul, Skanavi, Goldman, Barzakovskty, Bogoroditekiy and others), was dedicated to
the study of dielectric polarization and leases of polycrystalline dielectrics, pro-
duced as a result of reactions in the solid state occurring at high temperatures.
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
dioxide in the crystalline form of metallic rutile or tltanate of the second group
of Men.deleyev's periodic system.
The reason for the high dieLatric coefficient of rutile (TiO2), known from
1902, and perovakite (CaTi ), first established by Soviet physicists, was explained
by their theoretical work (Slmnavi).
In complex crystals, as in rutile ana peroval)ite, neither Born's nor Clausius-
Mosotti'e formulas car. be used to calculate the dielectric coefficient. Born's
formula is based on the fact that the local field is equal to the aveaege macro-
scopic one. In Cal sius-Mosetti's formula, the internal field should be equal t,.
the Lorentz field (41r I~ where I is the electrical moment of a unit volume).
Calculation of the internal field in the crystal lattice can be made in the first
approximation if it is based on the so-called "point" model of an ionic crystal
(point ions, which acquire dipolar momente due to the field). A method of cal-
culating the internal field, based. on Iorentz's method, leads 1.a the and to the
following general equation for electronic polarization in place of the Clauslua-
Mosotti's equation.
- _
mjaji? z oci a4 ('zkcEj+njej:4-7",~`jj-~?iGr4. ~ }O6%3] (1)
~7r J=~ J/
laiai~~~~1~k~ cj',cj) l-O~a3)
where e is the square of the index of refraction, N is the number of molecules
in one cu an of the crystal, a is the electronic polerizability of the Jth
ion, nj the number of j ions in tI'e molecule, of is the number of ions of different
geometrical distribution in the lattice, C , C k and J are the so-called
structural coefficients of the internml fi~ld, Lependunt only on the geometry of
the lattice and determined by the given pol#rizatillity of the ions and the given
external field supplementing the Lorentz internal field which 1s created b' the
polarized ions which surround the ion in question. 0(a 3) designates the small
terms having three and more multiples of the polerizability. If all the struc-
tural coefficients are equal to zero, then the above formula becomes the formula
of Clauuelus-Mosotti,
E_I _
E+2 3 ,N nJaJ
The -alculated structural coefficients for rutile and perovekite late ices
show that formula (1) in contrast to the Clausdns-16oboVti formula, is substan-
tiated fully by experiments. With infrared and lower frequesoies, it is neces-
sary to take into account, in addition to the electronic polarization of the'ions,
the dierlacemert of positive Ions relative to the ne tlve. In this case, for-
for rutile appears as follows, if the small terms are disregarded:
4 ?r N O(/ + 2 a2 -F- 0(i
w'nere a is the polarizability of the ionic displacement, relative to a titanium
ion. It is clear from (3), that even with the introduction of w all Obi, t
abraptly increases, since the numerator increases and the denominator decreases.
The physical nature of this effect consists of the fact that with Ionic displace-
ment there arises a large additional Internal field in the some direction as the
external field, which aide polarization. Because of this, a small polarirabillty
of ionic displacement (0k{ ) leads to a high value of dielectric coefficient.
I
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
STAT
It can be considered that the high dielectric coefficient of cryWals of the
rutile, and rutile and perovekite type, is greatly dependent In the first place on
the high electronic polarizability of the oxygen Ions. With a high density, We
results in a high index of refraction in conjunction with a large additional in-
ternal field, directed toward the external field, and with an abruptly rising di-
electric coefficient with the coefficient the change from light to infrared fre-
quencies ('Prom 7.3 to 173 for rutile), oven with emall ion displacement.
An outstanding achievement of Soviet physicists (Bul and Gol'dman) was the
discovery of the extraordinarily high dielectric coefficient ( F- equals 1,000 -
2,000), whicu passes through a sharp maximum when the temperature is increased,
and dielectric hysteresis of barium metatitanate (BaT103). Detailed investign-
ticme of the characteristics of this type of dielectric disclosed a new and very
interesting type of dielectric polarization. This polarization is similar 'try the
piezoelectric effect but is distinguished from it by a high value for a ix. a
very wide temperature range (from temperatures of liquid helium to plus 250 to
3000 C), The extraordinarily high dielectric coefficient of barium matatitanate
and also its sharp dependence on the temperature and the voltage of the electric
field open up many possibilities for various applications in electricity and radio.
The theoretical treatment or the polarization mechanism of barium metatitamate and
piezoelectric effects was later developed by Ginzburg. He showed that a phase
transition of the second order, occurring at a definite point (Curie point), can
cause the dielectric coefficient to tend toward infinity at this temperature, if,
in this transition, spontaneous polarization disappears. Roentgen's investigation
of the barium metatitamnte structure, at temperatures above and below the Curie
point, and measuring the dependence of the thermal capacity on the temperature,
showed that in barium metatitamate, at temperatures which correspond to the sharp
maximum E.(aroumd 800 C), the phase transition of the second order can take place.
Investigation of materials or the eystgm T102 - BaO (Skanavi)
shared that a change i'' the relation T1O2/BaO shifts the Curie point and
,we a the dielectric coefficient. In barium tetratitanate (1 T:02 . BaO) d - 30
sad changes very little with temmpereture, making It -.cssible to use it for tam-
perature-stable condensers. The work on the dielectric characteristics of rutile
and titamntes uncovered a wide range of practical products which were developed
during the war and which are being developed now. The ceramic condensers which
possessed these given characteristics were obtained by combining titanatee of
various crystalline structures, having different dielectric coefficients and tem-
perature coefficients. Rules were established and substantiated, by Roentgen and
other investigators, which determine the characteristics of the combined poly
crystalline material (Bul, Skanvi).
Dielectric losses in polycrystalline dielectrics, which were also studied
very Intensely, depend on the condition and quantity of the vitreous layer.
3. The work of Soviet physicists (Robeko, Suvshinakiy, Zhurkov, Shiehkin,
G. Mikhaylov and others) in the field of dielectric char?_oterlatice of amorphous
bodies was fruitful not only with respect to the theoretical explanation of the
processes occurring in an amorphous body, but also from the practical viewpoint.
Investigations on the relation of the lose angle and dielectric coefficient,
conducted over very wide temperature intervals for a whole series of sepereaoled
liquids containing polar molecules (glycerin, phenolphthalein, ieobutyl, alcohol,
etc.), showed that the lose angle passes through a clearly expressed temperature
m:.rtmrm. The7 also showed that the dielectric coefficient changes with tempera-
ture similarly to polar liquids. This was found at temperatures In which the eub-
stance was found In a solid--amorphous condition. An increase in frequency shifts
the temperature i y1tenn toward higher temperatures, which also takes place in
polar liquids. Recently it was shown (G. Mikhaylov) that in solid-amorphous
polymers containing polar groups the lose angle also has a second weaker tempera-
ture maximum In the low temperature region.
SLSanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
All this definitely substant_ates the possibility of rotating polar groups in
solid substances.
and gives the whole process tha evident character of dipolar loss.
constant of dielectric polarization was, evaluate4 approximately as 1/2r , where
temperature). The congruence between the indicated relationships brought to mind
the existence of the deep connection between dielectric and mechanical relaxation
The presence of this connection was substantiated by parallel measurements
of the dielectric coefficient and plastic deformation for amorphous bodies (Kobeko,
Huvehinakiy, and others).
Thus, one can assume that the Aisplacements and reversals of dipolar groups
are related to the displacements and reversals of whole molecular groups caused
by the plastic deformation.
Research on amorphous conditions made possible the development of ner high-
quality insulating materials. One of the first of such materials was the Soviet
polystyrene which has vory low dielectric losses (a lose angle of 0.5 - 1' at
high frequencies), very high volumetric resistance, and other valuable character-
istics, in particular ease of manufacture.
The study of dielectric and mechanical characteristics of rubber (Kobsko,
Ponomarev) led to the development of an unusual sulphur-free "ebonite-eslapon
vhioh is the product of polymerization of rubber and which has great thermosta-
bility, is easily workable, and has a comparatively small loss angle (1 - 2' at
high frequencies). This material in all respects surpasses common ebonite con-
taining sulfur. Each atom of sulfur is attached to 2 atoms of carbon and at the
same time, under the influence of carbon and hydrogen n cme is strongly polarized,
forming a permanent electrical moment. The mobility of the hydrocarbon chain in
rubber makes it possible to form a polar "sulfur bridge" co achieve the rotational
oscillations which occur in thermal agitation. The electric field orients the
"sulfur dipoles." The process of establishing this orientation at high frequencies
leads to clearly expressed dipolar losses. The absence of sulfur in eekapn and
the high degree of its polymerization appear to be the basic reasons for the small
dielectric losses of eakapcn.
Soviet scientists have been working for acme time in the field of they of
dielectric losses. By broad and systematic studies it was established (Iazarev
and others) that nebye's classical theory of dipolar losses is applied qualita-
tively to a majority of the amorphous commercial dielectrics above their solidi-
fication point (for example, oils, lacquers, compounds), but this theory does not
give a quantitative confirmation, experimentally, for the indicated dielectrics.
In connection with this, a series of o^.ewpoints were developed cm the mmobanism
of dielectric losses in various classes of dielectrics which has greatly Lid,,.
the development of new dielectrics with small losses.
Most of the work relating to the theory of dielectric losses in solid di-
electrics which has been published in foreign literature has a purely phenomenal
aft
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
character. The autao: prese_?ts the r- that aej .. the fuuction f..i~ current
drop with t"?',e end, using the known principle of eupe?poeition of currents in
dielectrics, with the aid of more or less exact and cumbersome calculations, ob-
tains an expression for E and tg 5 of the dielectric, depending on frequency.
Soviet scientists made attempts to develop theories of dielectric losses,
based on specific structural characteristics of the class of d.ielectric9 under
investigation.
4. The measurement of dielectric losses in crystalline dielectrics (Bogoro-
ditekiy and i+hlyehev) established that the dielectric losses in crystals have a
purely ohmic character (conductivity losses). On the other hand, in a series of
amorphous dielectrics, which do not contain polar groups and molecules, losses
were uncovered at high frequencies which were not of the conductivity type (the
work of Sobeko-Aleksandrov, Skanavi, Bogorcditakiy, .aljahev, zaidberg). Tae
presence of these losses is of great practical significance since they limit
greatly the use of a whole series of amorphous dielectrics (glass, chiefly) for
hie frenuenrv inaii1atioo. The work of Soviet physicists (Hobeko, Shanavi) An
studying the mechanism of dielectric losses in glass led to the discovery of a
new group of effects in solid-amorphous dielectrics which were similar to those
pror.uced by the rotation of polar molecules. This new group of effects is caused.
by the gradual shifting of weakly connected ions or charge groups.
The movement of such charged particles under a constant voltage brings about
a decreasing absorption current with time, and with voltage, brings about dielec-
tric losses of the relaxation type.
Phi absorption current in a solid dielectric with the indicated ionic move-
ment should, as shown by calculations, fall with time according to a simple ex-
ponential law, whereupon the time constant of the current drop is propourtional to
the relaxation time of the weakly connected ions. It is equal to T= ' y
where U is the activation energy, v -natural frequency of oscillation. However,
to describe the experimental data more adequately, we may use an exponential
function of the form I equals A(t-t- 9 )-n which reflects the process in the limited
time interval (A, 9 and n are :onstants).
This discrepancy is explained by the fact that in a real dielectric there are
a number of groups of ions with different activation energies, i.e., with different
relaxation periods, In addition, a number of other processes occur (e.g., accumu-
lation of space charge) whose combination complicates the law defining the drop in
absorption current.
The use of the exponential function of current drop leads to cumbersome mathe-
matical computations and does not help very much in explaining, the mechanism of di-
electric losses. The exponential function of the drop reflects one of the processes
(in many cases the prevailing one) existing in the dielectric. The theory of losses,
built up primarily on the use of this function, even though it cannot claim to be
rigid and complete, connects, as does Debye's for polar liquids, the values of the
dielectric losses in a solid-amorphous dielectric with its molecular constants.
The work of Soviet physicists, dedicated to. the use and development of the
theory of losses in solid-amorphous dielectrics (chiefly in glass) helped to
create new viewpoints on their mecharimms and, to a greater degree, contributed
to the discovery of new methods for obtaining glass with small dielectric looses,
A series of interesting facts were established (the work of %beko, Shaaavi,
) rtyuebov, Gladkukh and others) relating to theInfluence of the composition of
the glass on its dielectric losses. These facts conceYn: (1) the "neutralizing"
effect wherein the detrimental effect of one alkaline metal can neutralize the in-
troduction of the oxide of another alkaline metal (this neutralization is so great
that an alkaline glass can have as small a lose angle as pure glass without extra-
neous ions); (2) the "crystallizing" effect as a result of which crystallization
of the amorphous dielectric (sugar, glass) abruptly lowers its lose angle at high
frequencies, etc.
F'A
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
This work permitted the classification of dielectric losses in glass, which
can be cr.,pidered as consisting of three parts: (1) conductivity losses which
appear at low frequencies and high temperatures, (2) relaxation losses which ap-
pear at high frequencies and (3) structural losses, whose loss angle does not de-
pend on the temperature. The latter appear at high frequencies and low tempera-
tures. Their mechanism is not sufficiently clear as yet.
the dielectric properties of various types of materials has served as a theoretical
basis for obtaining new insulating materials with high insulating properties.
This concerns first of all the aOveliJPsteY{t of new inorganic materials (glass
and ceramic) whose application plays an especially important role is high frequency
techniques, since they show considerably smaller dielectric losses than many organic
materials, and very high thermostability,
The research conditions in our country are very favorable for complex projects
in which the activity of the physicists is in close coordination with that of the
chemists and the technologists. Electric insulation is the very fi ld of science
and technology where such cooperation is indispensable.
Ceramic materials exist in crystalline and amorphous phases. It may be that
the dielectric losses in ceramic materials are, in the first approximation, the am
of the losses in each of three phases.
As was indicated above, losses in crystalline dielectrics are this to conduc-
tivity, and therefore are very small at high frequencies. Consequently, the main
source of dielcotric losses In ceramic materials is vitreous interstratificatien.
The series of rules which has been established for dielectric loseea in glass
permit, a systematic approach to the selection of the composition and structure
of ceramic material with small dielectric losses. The principles for obtaining such
material are as follows: (1) the material must have a fine crystalline structure
with a minimum of vitreous interetratification; (2) the vitreous interetratification,
as far as possible, must not contain alkalies, or the effect of these alkalies must
be neutralized by heavy Lmetallic/ oxides.
In accordance with these general principles, a series of high-frequency ceramic
materials with small dielectric losses has been developed. The first material of
this group, produced in the Soviet Union and used widely, is a ceramic material
based on the mineral pyrophyllite (v-ineyev, Popov). A mixture of pyrophyllite and
clay possesses a number of properties that are of great industrial importance, in-
cluding great plasticity up to the *ime kilning takes place, which permits the use
of all methods of ceramic manufacture (machine molding, extrusion, pressing) which
have a long Lilning Interval.. In addition, ceramic materials with a pyrophyllite
base show, in comparison with ordinary porcelain insulators, small losses at high
frequencies (8 - 12 minutes at a frequency of --- 106 cycles).
By putting heavy Lmetalli2 oxides in insulating porcelain instead of feldspar
it has been possible to "enrich" it considerably and to obtain the so-called radio
porcelain, not inferior to pyrophyllite in dielectric properties and technologically
close to common porcelain (the work of Bogoroditekiy and Fridberg).
Work an obtaining ceramic materials with very small losees and other special
properties (e.g., a high dielectric coefficient and stability during temperature
changes--for condenser ceramics) has been carried out intensively.
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
very small lose anglesat high frequencies ()E a equals - These materials have
3').
Special attention has been given to the preparation of ceramic materials for
high-frequency eondenserm. Bogoroditskiy, and others, prepared "tikond," a ceramic
material based on titanium dioxide and clay with a high dielectric coefficient
2 equals 60 - 70 and a relatively large negative temperature coefficient of 14s
variation:
I J'-
d7 = S^10-9 11c%gree
This material is an appropriate dielectric i'or special condensers which oampeasatoa
for frequency variation during temperature change in radio circuits, due to the,
negative temperature coefficient of the eapaaitanam,
During the war the importance of high-frequency crystalline ceramic materials
increased even more, due to their low coot and abundant supply of the ore used to
produce them, and also the possibility of regulating the dielectric coefficient and
its temperature coefficient by varying the composition and structure of the crys_
talline phase of?these materials.
A ner ceramic material for compensating radio condensers based on calcium and
magnesium titanahas is "tidal" (Skansvi), with a still hlg er neiptive t rature
coefficient of the dielectric coefficient than tikond (1 : -10 - 10-'- 1/degree),
and a high dielectric coefficient ( a equals 70 - 80). ha lose angle for tidol is
2 - 3'. A series of materials based on magnesium titanates were developed (Vul,
Skanavi, Barzakovskiy, Bochkarev). These oeae40 materials were intended for radio
condensers of large and small roactiwe tower, They ahoy a very mull lose angle
equals 0.5 - 1' at high frequ.tnoies), which is largely independent of frequency
and temperature; a dielectric coefficient 2 equals 14 - 16; and, what is:eepeoiaily
important, stability of dielectric coefficient at high temperatures:
_ _ - (0.4 - 1.3) ? 10"4 1/degree
I dT
A material has been prepared based on titanium dioxide and clay with aluminum
oxide, called "tiglin" (81Molenekiy), also very useful in radio condensers. The
discovery of the possibility of regulating the dielectric coefficient and .ts tem-
perature coefficient in titanium ceramic materials by application of Liktenekvr's?
ekanavi) Law, is or
materials of different structure, It is rat Import m e~m~ oftn
etraaturee in the asking process by special methods. In these eases, it is possible
to calculate beforehand the dielectric coefficient of the combined material by the
dielectrin coefficient components:
? 1g e = z lg a1 + (l-z)lgE 2,
where E l and a are the dielectric coefficients of the components and he
z is t
volume concentra?ice of the first component,
respect LO temperature gives Differentiating this equation with
d -'x. T -f-.(1-x) 1 82 QT1
i.e., the temperature coefficient changes linearly with concentration.
By combining materials with positive and negative temperatures coefficients a
it is possible to obtain a given dielectric coefficient with a great degree of so-
curacy, and, what Is especially important, a given temperature coefficient of the
dielectric coefficient.
- 12 -
+w.
Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
STAT
By this method many condenser ceramic materialy have been developed with
various, temperature coefficients 6 from +1 10' to -12 - 10-4, molding
temperature coefficients E near zero (Skanavi, Stepanov, Voronkov, Bogoroditskiy,
Smolenakiy, odelevekiy, Nekraeov, and others).
Following the diLcovery of the unusually high dielectric coefficient of barium
metatitanate (Vul and Gol'dman), ceramic crystalline dielectrics based on barium
metatitanate were manufactured with a duper-high dielectric coefficient, i.e.,
from 700 to 1,500 at rein temperature (Skanavi, Voronkov, Odelevekiy). Work has
been done on the use of these materials in various fields of electrical and radio
engineering. Crystalline ceramic dielectrics have been produced based on barium
tetratitanate (4 Ti02 ? BaO) and similar compounds (Skanavi) with a very email
lose angle at high frequencies, a dielectric coefficient of about 30, ure
coefficient approximating zero, and a very high volume resistivity (1.015-10ib obm/cm).
These dielectrics' made it possible to manufacture thermostable ceramic radio con-
densers of high specific capacitance.
There have also been a number of achievements in the Soviet Union in the field
of application of glass as an insulating material.
First of all, there are the special types of glass used in radio engineering,
(e.g., for high-power radio tubes). Glass No 23 and Glass No 46 have small di-
electric losses, in addition to the necessary technological properties. Potassium
pyr,x (Bogaroaitskiy and Friberg) has been manufactured with a very small lose angle
at high frequencies and with thermostability as well.
Chernyak, Aslanova, and others have solved the basic pr-'lems involving glass
fibers and glass fabrics in the heatproofing of electrical insulation. Further
research is necessary here, especially in the selection of more suitable glans.
Andrianov, Tareyev, Chernyak, and other Soviet scientists have done consid-
erable research an the thermostebildty (heatproof qualities) and beat conduction
of insulation; furthermore, 6likhaylov, Bogoroditekiy, and others are studying the
bygroscoptoity of ' ilectrios.
The produoti%... and application of g]yptal lacquers have greatly increased;
coatings of these lacquers possess higher heatproof qualities than the coatings of
shellac and asphalt lepquers. in addition, the dielectric properties of llyptal
lacquers are of a c atively high order.
Asphalt lacquers, which are replacing scarce shellac, are again being vilely
produced and employed in industry, especially for use in electrical machines.
Among the new plastics, those deserving greater attention are polystyrene
and eshapon, already mentioned. Andrianov and Gribanova have successfully developed
primarily new electrical insulating materials m the basis of silicone cemponnde.
These materials, liquid and solid, combine within themselves the properties of or-
ganic compound@ (plasticity, elasticity, etc.) together with high thermoetability
(up to 3000 C). The indicated materials were developed as a result of extenbive
studies in a new branch of science, nenely the chemistry of silioopes (by Andrianov).
This short account does not exhaust the list of developments seen iii recent years
in the field of new insulating materials.
We will now ermine the studies of Soviet physicists on dielectric rupture.
The puncture voltage for insulators depends on the electrical stability of the
insulating material and, to the some extent, on its structure, which determines
the distribution of the electric field in the asulation and the ocudit~ons of
beat developing. Rupture should be studied not in models alone, but under prac-
tiual operating conditions.
- 13 -
Sanitized Copy Ap
proved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
I Mi
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
In the Soviet Union the theory -f rupture was begun 20 years ago (Semenov,
Fok, Bragin, and others). Broad experimental studies established the factors
characterizing thermal and electrical rupture of insulation. Criteria were
established for classifying rupture in each egperate case, in a series of di-
Fok, Semenov, and others first gave a strict theory of thermal rupture which
was successful in practice.
down or puncture voltage, which in thu case of thermal rupture, can be treated as
that maximum voltage (critical voltage) beginning with which the stationary state
is impossible. When the critical voltage Is attained, continuous heating begins
in the dielectric, which after continuous use for some time leads to thermal rupture.
The second group determines the "time of rupture"; that is, the time it takes the
dielectric to break down after reaching the critical voltage. Both groups of prob-
lems are of very great engineering interest because:
a. The calculation of the critical voltage of thermal rupture aloes it
possible to select properly that quality and construction in the insulation which
assures a full useful life.
b. The calculation of the "time of rupture" at voltages higher than the
critical permits one to calculate, fo., any given selection of the insulation, sta-
bility versus short-tine oi:ieas-voltage. The first group u`sae been ssolvedanti the
results have been used for a number of years. The second group has been solved
basically in recent years.
Thermal rapture in a dielectric is due to the fact that the generation
of hest from dielectric lose (or conductivity at constant voltage) increases as
the dielectric in heated. For most dielectrics, the liberation of heat depends
upon temper Lure in a limited interval according to the following exponential law:
Q : Qo ,a'? (T To)
where T and To are temperatures and a.is the temperature coefficient- of conductiv-
ity or the tangent of the angle of loss.
?p a= KaT+ Q.
t
where c in the specific hqat of the medium, p is the density, K is the heat
conductivity and q : -r3fO is the specific amount of heat liberated ( 7 Is the
conductance and S is the field intensity) or for the uniform case often not in
practice:
ar a'T+
at 6Kaz Q
In the solution of the first group of problems (of finding the critic l puncture
voltage), only the equilibrium state is examined ( a t 0 ) aid hence time is
naturally excluded from the examination.
tizeopy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
 Sanitized Copy Approved for Release 2011/09/19: CIA-RDP80-00809A000600270019-1
Th* differential equation
K" I'* T+ Q==,4
serves basically to calculate the critical voltage.
to the nonequilibriua state (during increases in the applied voltage)., which ap-
where B is the field intensity and T. is the maximum temperntuie of heating
for a given voltage.
When 25
1, > 0, -re have the equilibrium state; and when we
,
have the nonequilibrium state of the dielectric. The equation 0 thus ap-
voltage of thermal rupture. For example, for a uniform alternating field intensity,
where E and. b are the dielectric permeability and angle of Tee, respectively,
of the dielectric at ordinary temperatures; f to the frequency; p (C) Is
a rather complex function, whose approximate value is
C= X' ;kA
X (/