SCIENTIFIC ABSTRACT MILLER, M. A. - MILLER, M. YE.

• Document Type: CREST
• Collection: Scientific Abstracts
• Document Number (FOIA) /ESDN (CREST): CIA-RDP86-00513R001134310010-7
• Release Decision: RIF
• Original Classification: S
• Document Page Count: 100
• Document Creation Date: January 3, 2017
• Document Release Date: July 31, 2000
• Sequence Number: 10
• Publication Date: December 31, 1967
• Content Type: SCIENTIFIC ABSTRACT


KILUIR, N.A.
-w~W~tAUVWKI$AOLWW46-
Surface electromagnetic waves in rectangular channels. Zhur.tekh.fis.
25 no.11:1972-1982 0 $55. (NmA 9:1)
(Wave guide*) (Ilectric waves)
-,~e

USSi/Radiopbysics Superbigh Frequencies, I-11
Abst Journalt Referat Zhur - Fizika, No 12p 1956.9 354p
Author% Bespalov., V. I., Miller, M. A.
Institution:
Titlet Electromagnetic Waves in Rectangular Slots in Which the Bottom
Is Covered by Dielectric
Original
Periodical: Uch. zap. Gorlkovakoun-t., 1956, 3o, 61-75
Abstract: A discussion of the prcpagation of electromagnetic waves in a rec-
-taugular U-shaped slot, the bottom of which is covered *ith a layer
of isotropic dielectric. A now method is proposed for knding the
natural waves, propagating along the slotj the fields are found in
the form of a superposition of TZ axd TH waves relative to the di-
roction of the aperture of the #lot. From the dispersion equation
obtained it follows that the attenuation factor of the f;Old of the
surface wave is independent of the width of the slot and consequently,
this dispersion equation is valid also for a slot that varies in
Card 1/2

USSR/Rullophysics - Superhigh Frequencies., I-11
Abst Journals Referat Zhur - Fizika, No 12, 1956, 35453
Abstracts width along the direction of propagation. Usual methods are used
to obtag*the attenuation due to the losses In the metal and in
the dielectric, and to find the directivity pattern of the radia-
tion of the first propagating wave from the aperture Ct the slot.
The directivity pattern for this wave has a trough-like form.
Bibliography., 9 titles.
Card 2/2

? ~:~i ic
i(. i Y , L . !". . , "'.1 - -I. . 1 1;, -, - )
AVEF=V Y,.Lll. 7. A.
r L
V Kr
tbe Thr*e.Cqutiw"r ftow 'a
',x DIOdo Iloise Gebemtor in M. Bravo :.ZhJVCtA>Vl~ kiy , A.
V
U T ;.rPkhL)V4. so 6.

O'B JVC T USSR / PHYSICS CARD 1 / 2 PA - 1832
AUTHOR 1K , TALANOV,V.I.
.TITLE Electromagnetic Surface Waves directioned by a Boundary with a
Slight Curvature.
.1
;FERIODIGAL Zurne'techn.fis, z6f fusc.12, 2755-2765 (1956)
Issued: 1 / 195-7
In the paper by the same authors, Zurn.techn.fis.21, fasc.11, 1610 (1955) the
properties of electromagnetic surface waves directioned by flat boundaries were
investigated. This problem may be looked upon as a limiting case of the problem
concerning waves directioned by a cylinder (in the case of an infinite value of
the ratio between the cylinder radius r 1 and the wave length A ). The present
work aims at investigating this boundary transition more closely than had been
the case in the previous work. It was furthermore important to evaluate the dis-
tortions which were carried into the surface field (in the case of high but
finite values of the curvature radius of the directioning boundary). Apart from
the practical point of view, this is of interest also as a matter of principle,
because there exists a certain class of waves which in the vicinity of a direc-
tioning boundary lose their surface character even if the curvature be ever so
small. At first the equation for the wave numbers is set up.For this purpose
a cylinder with any radius r-r 1, on the surface of which homogeneous boundary
conditions prevail, is investigated. The problem consists in finding the radicals
of the equations which were set up. With any value of the parameter p-kr 1 these
equations can be solved only numerically; if p, - kr, >> 1 (r, >/-> ~ ) is

V
Zurlttechn.fiap26, faso.12, 2755-2765 (1956) CARD 2 / 2 ?A - 1832
sufficiently great, the asymptotic value of the radicals can be found analyti-
cally. (k-. 2") The condition p, ~> I corresponds to the case of alight
curvatures of tye"directioning boundary, which interests us here. Only those
boundaries are investigated here which are curved only in the direction of the
propagated waves. For this reason the azimuthal waves were dealt with. Azimuthal
waves can be divided into two groups: stable waves, which remain on the surface
in the case of any curvature of the boundary, which are threedimensional, and
may e.g. be realized in the grooves of a rectangular profile with ideally con-
ductive lateral walls, and unstable waves in the case of which, even if the
curvatures of the boundary are alight, a radiation field occurs, and which are
in the main line two-dimensional. On the basis of the solution of the problem
of the rotation of azimuthal waves, corrections to the propagation constants
of the waves, which are characterized by a finite but sufficiently small
curvature of the directioning surface, are found. For nonstable waves the pro-
pagation constant becomes complex. Radiation losses diminish considerably in
the case of a curvature of the boundary in the direction of the field.
INSTITUTION:

64. ON "M WWR" ZQUATUW POP. Moos W Tag
~
LRY Or WMTALLIC ARRIAU -IAAAW-SL.-
M MY Z-70,27k706956). In wasam.-
'a Is,, Vol. 36, No.
Diwussion: of an error in the use of integral equations In tM
Wory of aerials when the field due to the exagnetle currents Is
Ignored. Sims an arbitrary field can be represeented as a field of
Vurs17 slactric currents distributed over a aimed surface Z, the
Introduction of fictitious Mag"tic currents does not appear
RecessAn. It Is, however, a Coure"Ient Madient wwn calculamig
ths trapedance dwwtartsilca of these aerials, For the cw* of An
serial under load or for a finite conductivity of the ruetal, the
tangential consr4nents of Lhe electric and T!M-Lk~- over Z are
related by the roUtion RT - f(HT) aj Z.F.Voyner
-51"

MILLER, M.
Tensiometryj a modern measurinF, method. p.165.
(Elektrotechnik, Vol. 12, No. 6, June 19157, Praha, Czechoslovakia)
SO: Monthly List of East European Accessions (EEAL) 1E. Vol. 6, Nc. 9, Sept. 1957. Uncl

o6507
AUTHOR: Miller, M.A. SOV/141-58-4-24/26
TITLE*. Princ p -e.6'f Generation of High-Frequency Oscillations
(Ob odnom printsipe generatsii vysokochastotnykh
kolebaniy)
PERIODICAL:Izvestiya vysshikh uchebnykh zavedeniyj Radiofisika,
1958, Nr 4, pp 166-167 (USSR)
ABSTRACTz It was shown earlier by the author (Ref 1) that the
motion of a charged particle in a slightly non-
homogeneous electromagnetic field can be represented
In the form of the super-position of an oscillatory
motion with a frequency w and an average motion
(averaged over a period 211/w) which can be represented
R = -1C 40; R describes the motion as a function of
time while 0 represents the high-frequency potential.
The sum of the kinetic energy of the averaged motion and
the mean kinetic energy of the oscillatory motion in
constant in such a system. Consequently, if a beam of
particles is directed towards an increasing potential
Card 1/3 at the point of reflection corresponding to I = 0. a

o65o8
SOV/141 _58-4-24/26
A Principle of Generation of High-Frequency Oscillations
Card 2/3
total transformation of the energy of linear motion
in the oscillatory energy takes place. Thus, assuming
that during the instant when the total velocity is
zero and the particles recede from the interaction
space, the kinetic energy of the particles becomes
transferred into the field. Consequently, the principle
can be used to devise a high-frequency oscillator.
An example of this type of oscillatory system is
considered. The oscillator contains a parallel tank
with a capacitance C and an inductance L, the quality
factor of the system being Q. It is shown that the
system can be described by Eq (3), where Cl is a
certain additional capacitance due to the presence of
the space charge. The author expresses his gratitude
to A.V.Gaponov, Ye.V.Zagryadskiy and M.I.Kuznetsov for

06508
SOV/141-58-4-24/26
A Principle of Generation of fligh-Frequency Oscillations
a number of valuable remarks. There is 1 Soviet
reference.
ASSOCIATION: Issledovatel'skiy radiofizicheakiy institut pri
Gorikovskom universitete (Radiophysics Research
Institute of the Gorlkiy University)
SUBMITTED: 14th June 1958
Card 3/3

Z&
/411 A A
AUTHORS: apCODY-1, A. V., Miller, M. A. 56-1-44/56
TITLE% On the Potential Wells for Charged ?articles In a High-
-Frequency Slectromagne 'tic Field (0 gotentsiallnykh yamakh
dlya zaryazhannykh chactito v vysokochastotnom elektro-
magnitnom pole)
PERIODICALt Zhurnal Zkoperimentallnoy i Toortticheakoy Fizlkil 1958#
Vol. 34, lir 1, pp. 242-243 (USSR)
ABSTRACT: As is well-known there exist no absolute maxima and minima
of the potential in an electromagnetic field in solenoidal
domains, which excludes the possibility that a charged
particle remainn in the state of stable equilibrium. This
fact also prevents the possibility of the localization of a
particle, provided that under localization a state to under-
stood ih which a particle with an energy staying below a
certain limit can leave a limited domain under no initial
conditions whatever, This statement, howeverg does not apply
to the case of a higlt-frequc-ncyoltctrofqagnetic field where
the particle (as shown here) can be !Qcallaed. The authors
investigate a particle with the ohar~;e e and with the mass
Card 1/3 m which moves in the outer electroma,6-nctic field

On the Potential Wells "or Charged Particles in a Hi.-h- 56-1-44/56
wFrequencj Electromagnetic Field
iLit --I, iwt
9 (r,t) - E H(r)e In nonrelativiatic
(r)e H(r,t)
approximation the equation of moti-:n reads r +
+(1/c) rr where 7Z . e/m applies. At a sufficiently
hiGh frequencycoof the outer field the solutions of the
just-menti,)ned equation can be represented in the rorm of
a sum of a function ~'(t)_-Iowly varying (vrith regard to the
period of the oscillaiion's of the outer field) and of a
function i~`(t) oscillating with the frequencyco. After
averaging Co above-nentioned equation over the period of
the high-frequency field the follovine p1lation is obtained
for i.,/M: Y (t) = In, 4) - (~1124 2 1EI . B~, averagina over
the t?mp the Porc'e act j3 upon the particle beoioier; a
potential forco, where the potentIRI of the force is pro-
portional to the square of the molulus of the electric field
strenth and is not dependent on the si&n of the charge.
There exists an infinite number of possibilities for the
construction of the potential Wells for (1). The simpleat
of them is realized in the quasielectrosTatic multipole
fields. rn order to determine the natiire of the motion of
the particle within the potential wells the authors
Card 2/3 investigate the first integral of the last-mentioned

On the Potential Wells for Charged Particles in a High-FrequencY 56-1-44/56
Electromagnetic Field
equations. When E - 0 applies in the center of the potential
well, the particles with an energy of - V are localized
within a certain domain on whose boundarigs the conditions
LcJ2 /h7j >> I E I > 2.(V 1) 112 are valid. It is also possible
to build up threedimenNonal potential wells of unidimensional
and two-dimensional potential wells. There are 3 referenc*s,
2 of which are Slavic.
ASSOCIATION: Gorlkly Oftin Udiarsity . (Gorlkovskiy gosudarstvennyy
universitet)
SUBMITTED: October 15, 1957
AVAILABLE: Library of Congress
Card 3/3

AUTHORS: Gaponov, A. V., Miller, I A SOV/56-34 A;
TITLEt On the Use of Moving High-Frequency Potential Wells for the
Acceleration of Charged Particles (0b ibpollzovanii dvizhush-
chikhsya vysokochastotnykh potentaiallnykh yam dlya u3koreniyP_
zaryazhennykh chastits)
PERIODICAL: Zhurnal Eksperinientallnoy i Teoreticheskoy Fiziki, 1958,
Vol. 34, Nr 3, pp. 751-752 (USSR)
ABSTRACT: When using oscillations of different frequencies generally a
potential relief f(,?O,t) changing with increasing time is ob-
tained. This way especially an accelerated motion of potential
wells can be realized and consequently charged particles local-
ized in such wells can be accelerated. The authors investigate
2 wave running in opposite directions (+z). W'th equal frecluenc-
~-Y (x 1W't
ies and amplitudes they form a standing wave 0 Y,Z)e , where
%E+(x,y,z) is a real function. The potential corresponding to
t9is field may give absolute minima."For the reason of a dis-
placement of the potential wells on the z-axis the phase of one
of the oppositely running waves must be changed. The authors re-
Card 1/3 strict themselves to a non-relativistic motionlAcj I rcpT (1.1)
where E is the length of the gap,
T is the period of oscillation of the control field and
cp is the average velocity of the particle (electron).
For the purpose of analysis it is assumed that the electro-
magnetic field inside the gap is in the form of
iWt iWt
E(r)e and H(r)e The motion of an electron can
therefore be represented as the super-position of an
Card 1/6

86858
S/141/60/003/005/013/026
E192/E382
Some Properties of Electron Gaps with Large Transit Angles
oscillatory motion:
(1) 2 iWt
r (t) q/W E(r)e
~(I)M iTi/wE(r)eiWt
and a continuous motion obeying:
..(0)
r = 'V~P(r(O))
(2.1)
(2.2)
where I is the ratio of the electron charge, e , to its mass,
is the high-frequency potential expressed bys
� = JIF,/2wl2
(2-3)
Card 2/6

86858
S/141/60/003/005/013/o26
E192/E382
Some Properties of Electron Gaps with Large Transit Angles
This averaging description of the motion is valid if the
conditions defined by Eqs. (2.4), (2.5) and (2.6) are met.
on the basis of Eq. (2.2), the integral of the averaged
energy is given by:
;. (0)2 /2 + (t (r(0)) = const
Firat, the ideal case is ccvndAdered which corresponds to
the maximum value of the electron conductivity G )J1
(Refs. 7, 8). A rectilinear beam of particles with input
velocity v BX enters a gap whose high-frequency potential
increases monotonically. It is now possible to choose the
"output surfacet' of the gap so that the o'verall velocity of
the particles on it is zero (see Eq. 3-1). In this case, the
kinetic energy of the electrons will be fully tranar4kre4:
to the electromagnetic field. This corresponds to the
maximum possible value of the electron conductance. However~
Card 3/6

86858
s/i4l/60/003/005/013/026
E192/F,382
Some Properties of Electron Gaps with Large Transit Angles
in actual conditions the value of the electron conductance
will be lower. The optimum Y2 lue of the conductance is given
by Eq. (4.4), where V. = mc/p_ , I is the current in
the beam and f(r) is the field distribution in the gap.
In the case of a non-optimum bunching or sorting of particles
it is necessary to derive special formulae. The beam is
assumed to be rectilinear and it is oriented at an angle
a%I with respect to the output boundary x = 0 of the gap.
It is assumed that the tangential component of the vector E
at this boundary in equal to zero so that E is expressed
by Eq. (3.1), where V.,,.f(zn )/LEis the amplitude of the
uniform field in the bunching space. on the basis of
Eqs. (4.1) and (4.2), the trajectories of the electrons can
be written as Eqs. (5.2) and (5-3), where e = wt and
p is the phase bunching parameter which is defined by
Card 4/6

86858
s/i4i/60/003/005/013/026
E192/E382
Some Properties of Electron Gaps with Large Transit Angles
Eq. (5-4); here, t b = 'rb/w is the instant of the particles
entering the sorting space and xb and z b are the
coordinates determining the boundary of the sorting space.
Now, the electron conductance can be expressed as
G = GC)VI-T F(u) , where Ii is defined by Eq. (5-12). The
function F(p) can be referred to as the phase debunching
factor. At p > I it is impossible to obtain real values
f or and T' 3 In the limiting case, when ji .. )p 0 and
It: Ir , the function F(p) 4 k + I On the other
2 38mall
hand, for /p the function F(A) is approximately expressed
by Eq. (5-14). For the case when the field E in the
sorting or bunching space is in the form of Eq. (5-1) but
the output plane of the gap is at an angle y with the plane
x = 0 , the phase debunching factor is expressed by.,
Eq. (6.4), where ';r is defined by Eq. (6.2). For small
Card 5/6

86858
s/141/60/003/005/013/o26
E192/E382
Some Properties of Electron Gaps with Large Transit Angles
Eq. (6.4) is written as Eq. (6*5). A graph of the debunching
factor (given by Eq. 6.4) is shown in Fig. 1. It is of
some interest to determine the change in the velocity of
the electron beam while passing through a gap. The velocity
of the electrons in the direction of the field can be fdund
directly from Eqs. (5-3) and it is expressed by Eqs. (7-1).
The above analysis can easily be applied to the systems whose
gaps are subjected to the interaction of static fields.
There are 2 figures and 8 references: 5 Soviet and
3 English.
ASSOCIATION: Nauchno-issledovatel'skiy radiofizicheskiy
institut pri Gor,kovskom universitete
(Scientific Research Radiophysics Institute
of Gorlkiy University)
SUBMITTED: may A, 196o
card 6/6

191 :~_158ar
1J, Al -j,.o 0
to AUTHOR: _-Mille-
-
TITLE: Electron Gaps with
Oscillator Systems
s/i4i/60/003/003/014/026
E192/E382
Large Transit Angles in
PERIODICAL: Izvestiya vysshikh uchebnykh zavedeniy,
Radiofizika, 1960, Vol. 3, No. 5, pp. 848 - 859
TEXT: The article is a direct continuation of the preceding
work (see pp. 837 - 847 of this journal) and it is devoted
to the investigation of the interaction of thin electron beams
with electromagnetic fields in resonators. It is assumed that
an electron gap is in the form of a high-quality resonator
with a predetermined structure of the electromagnetic field
which corresponds to one of the natural frequencies of the
system. The field excdted in the resonator by the electron
beam oscillating at a frequency w is written as Eq. (2.1),
which is analogous to Eq. (5.2) of the preceding article.
Here, V-.,(t) is a slowly varying voltage function which is
described by Eq. (2.2) and LE is the characteristic linear
Card 1/6

86859
S/141/60/003/005/014/026
E192/E382
Electron Gaps with Large Transit Angles in Oscillator Systems
dimension of the resonator; f(r) in Eq. (2.1) represents
the distribution of the electriT field inside the resonator.
For the amplitude function V (t) it is possible to obtain
the following equation:
d2V dV, 2 2 da
-~j + 2iw- + (W - W iwa (2.4)
2 dt co6 V11 dt
where w co6 is the natural frequency of the resonator.
The simplified equations for the system are in the form of
Eqs. (2.6) and (2-7), where Q is the quality factor of the
resonator, while the parameters a s
Eqs. (2.8) and (2.9). The current
expressed by Eq. (2.10) so that a
by Eq. (2.11). It is now necessary
(3.2) of the preceding article with
Card 2/6
and a c are defined by
vector in Eq. (2.9) is
of Eq. (2.8) is given
to combine Eqs. (3-1) and
Eqs. (2.5) and (2.7).

86859
s/141/60/003/005/014/026
E192/E382
Electron Gaps with Large Transit Angles in oscillator Systems
For this purpose, the current of Eq. (2.10) can be represented
by Eq. (3.1). Now, the parameter a is expressed by Eq. (3-2)
where various constants are defined by Eqs. (3-3), (3-4) and
(3-5). The final system of simplified averaged equations
is in the form of Eqs. (3.6). Here, the quantity M determines
the frequency displacement due to the electron beam. It is
seen that for T1 = const., the frequency shift is given by
Eq. (3-7). Since the quantity 31' is comparatively small,
Eqs. (3.6) can be approximately written as Eqs. (3.9), where
Gpe3 is the effective conductance of the resonator and GN,
is the real component of the electron conductance. The
formulae are used to analyse the system with a two-dimensional
electron beam of finite thickness (D,X x ~ D which is
2)
parallel to the plane x = 0 In this case, the electron
conductance is given either by Eq. (4-7) or Eq. (4.8). In
particular, when D 1 = 0 , i.e. when there is no gap between
Card 3/6

86859
s/14i/60/003/005/014/026
E192/E382
Electron Gaps with Large Transit Angles in oscillator Systems
the beam and the collector, the electron conductance is defined
by Eqs. (4.9), where G I = I/Ve - The dependence of the
electron conductance for RD 2 = 1 on V, /Ve is illustrated
in Fig. Ia. If it is assumed that an ideally nar-row electron
beam enters the electron gap at an angle ab (with respect
to the collector plane), the electron conductance is defined
by Eq. (5.7). This formula is identical with Eq. (4.9),
if it is asrumed that Di = D2 = D . A graph of the electron
conductance expressed by Eq. (3-7) is shown in Fig. 16,. For
the case of a system with longitudinal bunching (an
inverted coaxial diode, where the electrons with the initial
velocity v b are injected through an external sheath having
a radius b ) the electron conductance is expressed by
Eq. (6.2). A graph of this function is given in Fig. 1B.
Card 4/6

86859
s/141/60/003/005/Oi4/o26
E192/E382
Electron Gaps with Large Transit Angles in oscillator Systems
It is of interest to estimate the principal parameters necessary
for obtaining the oscillation conditions in the above systems.
These parameters are the current I and the voltage V. I
corresponding to the steady-state oscillations in the system.
It is shown that the current is expressed by Eq. (7.2), where
L;r is the effective dimension of the resonator which is
W
defined by Eq. (7.3). The voltage VI is expressed as
follows:
V (kLE )2 (kLE)2
VI = --2- - _ 10 (volt) (7.6)
2
Ue n n
where n = At/T A is the transit time through the
gap having a length LE and T is the period of oscillations;
Card 5/6

86859
s/141/60/003/005/014/026
E192/E382
Electron Gaps with Large Transit Angles in Oscillator Systems
n thus represents the number of waves or oscillations along
the gap.
There are 2 figures and 9 Soviet references; one of the
references is translated from English.
ASSOCIATION: Nauchno-issledovatellskiy radiofizicheskiy
inst�tut pvi Gorlkovskorn universitete
(Scientific Research Radiophysics Institute
of Gorlkiy University)
SUBMITTED: Play 15, 1960
Card 6/6

3319? L - -
S/1 1/61/004/005/001/0'
E039/El35
AUTHORS: Miller, M.A., and Talanov, V,I.
TITLE: The use of the surface impedance concept in surface
electromagnetic wave theory. (Review)
PERIODICAL: Izvestiya vysshikh uchebnykh zavedeniy,
Radiofizika, v.4, no-5, 1961, 795-830
TEXT: This is a comprehensive review paper which deals wifli
some general questions on the way in which the theory of surface
electromagnetic waves is related to impedance and on the gui,lji~g
properties of boundaries, It is assumed that in the general
case surface impedance may possess spatial dispersion. The
value of this in the study of free waves, as well as for the
solution of the problem of surface field excitation by means of
various sources, is shown. The work is discussed under four
main headings, as follows.
1. Free surface waves, This section is divided into ten parts
and starts with a discussion on surface impedance, In the case
of a closed boundary surface the tangential form of the vectcr
field is given as:
Card 11.4

33199
The use of the surface impedance s/141/61/004/005/C)0-1/021
E039/E135
2
Ei Z ik bLl I k
k=1 LJ
where n Is normal to the surface, ik is the characteristi,-~
index oi orthogonal coordinates In the direction of tho ..' "!I
The tensor Z R 4. jX in a practical rationalised E T -'.' 41
ik = ik ik
of units (used in this survey) has the dimensions of impedan'r
and is called the surface impedance tensor. It is shown that
surface waves guided by a plane boundary become plane
heterogeneous waves and, for cylindrical surfaces, cylindr4-cdl.
heterogeneous waves, A large part of the work on surface wa%-F~s
is devoted to the guiding properties of surfaces.. The basic:
properties of surface waves are discussed in detail, firiefiv
the eondition for the existence of surface waves near a plane
z = 0 leadE to the relation:
Rey Imy -> 0
Card 2/1~'

33199
The use of the surface impedance S/141/61/004/005/001/021
E039/E135
The phase velocity of such waves is always less than the velocity
of propagation (for a two dimensional plane the velocity of light
c = I/ Ve-ji ). Discarding non-essentials the solution of the
equation for local fields for ~2 I gives:
Y� N (1-15)
if M and N are considered real it is comparatively simple to
classify all possible forms of surface_waves. Fig.1 shows five
different regions for the parameters M and N. In the first
three the condition (1.13) is satisfied. in region I (N > 0)
there is only.,one positive root (-_,, > 0, 0) and only one
Y
type of surfalce wave is possible. In r-3gion II,
(W > -N, N < 0, jj > 0) the simultaneous existence of two
propagated waves is permitted. In region III
- 2
(M 4C -N, N < 01 M >0) there are two complex roots with
positive real parts. The regions IV and V correspond to the
propagation of non-localised fields. In IV Orl < -N, N < 0, Irl 4 0)
Card 3/ 1~

33199
The use of the surface impedance s/14l/61/004/005/001/021
E039/EI35
both roots of Eq.(1.15) are complexv and in V (~I>-Nj N