Note: Descriptions are shown in the official language in which they were submitted.
~ -2-
~; 1330827
Background of the Invention
1. Field of the Invention
The present invention pertains to the production,
man$pulation and exploitation of high electrical charge
den~ity entities. More particularly, the pre~ent invention
relates to high negative electrical charge density entities,
generated by electrical di~3charge production, and which may
be utilized in the transfer of electrical energy.
2. Brief De_criDtion Or Prior Art
Intense plasma discharges, high intensity electron
beams and like phenomena have been the sub~ects of various
studies. Vacuum Arcs Theory and Application, Edited by J.M.
Lafferty, John Wiley & Sons, 1980, include~3 a brief history
of the study of vacuum discharges, as well as detailed
analyses of various features of vacuum arcs in general.
Attention has been focused on cathode spots and the erosion
of cathodes used in producing discharges, as well as anode
spots and structure of the dischargas. The structure of
electron beams has been de~cribed in terms of vortex
filaments. Various inve~3tigatorc3 have obtained evidence for
discharge ~tructures from target damage studies of witness
plate records formed by the incidence of the discharge upon
a plane plate interpo-3ed in the electrical path of the
discharge between the source and the anode. Pinhole camera
apparatus has also disclosed geometric structure indicative
of localized dense sources of other radiation, such as X-
rays and neutrons, attendant to plasma focus and related
discharge phenomena. Examples of anomalous structure in the
context of a plasma environment are varied, including
lightning, in particular ball lightning, and sparks of any
kind, including sparks resulting from the opening or closing
of relays under high voltage, or under low voltage with high
current flow.
The use of a dielectric member to constrain or guide a
high current discharge is known from studies of charged
particle beams propagating in close proximity to a
dielectric body. In such investigations, the entire
particle flux extracted from the source was directed along
,
. .
~ :
_3_ 827
the dielectric guide. Consequently, the behavior of the
particle flux was dominated by characteristics of the gross
~i discharge. As used herein, ngross discharge" means, in
- part, the electrons, positive ions, negative ions, neutral
' 5 particles and photons typically included in an electrical
discharge. Properties of particular discrete structure
present in the discharge are not clearly di~ferentiated from
average properties of the gross discharge. In such studies
utilizing a dielectric guide, the guide is employed wholly
for path constraint purposes. Dielectric guides are
utilized in the context of the present invention for the
manipulation of high charge density entities as opposed to
the gross discharge.
The structure in plasma discharges which has been noted
by prior investigators may not reflect the same causal
circumstances, nor even the same physical phenomena,
pertinent to the present invention. Whereas the high charge
density entities of the present invention may be present, if
unknown, in various discharges, the present invention
discloses an identification of the entities, techniques for
generating them, isolating them and manipulating them, and
applications for their use. The technology of the present
invention defines, at least in part, a new technology with
varied applications, including, but not limited to,
execution of very fast processes, transfer of energy
utilizing miniaturized components, time analysis of other
phenomena and spot production of X-rays.
i
:, .
..
: .
13430827
Summary of the Invention
The present invention involves a high charge density
entity being a relatively discrete, negatively charged, high
density state of matter that may be produced by the
application of a high electrical field between a cathode and
an anode. I have named this entity ELECTRUM VALIDUM,
abbreviated "EV," from the Greek "elektron" for electronic
charge, and from the Latin "valere" meaning to have power,
to be strong, and having the ability to unite. As will be
explained in more detail hereina~ter, EV's are also found to
exist in a gross electrieal discharge.
The present invention includes discrete EV's comprising
individual EV's as well as EV "chains" identified
hereinbelow. It is an ob~ect of the present invention to
provide for the generation of EV's within a discharge, and
for the separation of the EV's from the diffuse space charge
limited flux produced therewith.
It is a further object to manipulate EV's in time and
space.
It is yet another ob~ect to isolate and manipulate EV's
to achieve precise relative time interval control and
measurement.
In general, and according to the present invention,
EV's may be produced utilizing a generator such as, but not
limited to, a vacuum or gaseous diode. In one ~orm of such
a generator, dielectric material is disposed between an
emissive cathode and a second electrode, or anode, which is
thu~ ~hielded by the dielectric member from the cathode to
avoid direct cathode-to-anode discharge. The dielectric
member, however, provides a surface along which an EV may
move toward the anode. Such a dielectric member may be
constructed to define guides, such as channels or the like,
to constrain an EV to a defined path. A counterelectrode
may underlie the desired path on the opposite side of the
dielectric material to further constrain the EV to the
path. Adding a low pressure gas above the dielectric
surface facilitates ~ovement of the EV across the
dielectric.
, '
~"
15330827
In another form of generator, a cylindrically-symmetric
cathode is diqplaced from an anode over a gap which may be
in vacuum or qub~ect to low pressure gas within a dielectrlc
enclosure. In a variation of such struoture~ a cathode
5constructed on the exterior of a conical dlelectrlc member
having an anode positioned therewithin, may produce an EV
which can be launched acrosq a gap in vacuum or low pressure
gas, attracted by a counterelectrode carried on the outside
of a tubular dielectric member into which the EV is
10manipulated, as one form of an EV launcher.
Various cathode qtructures are provided, including
cylindrically symmetric as well aq planar, and techniques
for wetting the structures with conducting material to
repair erosion are also disclosed.
15A counterelectrode, positioned behind a dielectric
qtructure having an acute edge generally intermediate a
cathode and an anode in the càse of either a cylindrically-
symmetric generator or a generator for propagating EV's
along a surface, may be utilized to separate a desired EV
:- 20from electrons and ionq that may be presented with the
discharge by which the EV is formed. A similar structure
permits the selection of EV's from a multiple EV production.
The dielectric guide principles are further refined to
- provide devices whereby EV paths may be split, which even
25allows for the oontrolled arrival time of EV's at specific
points. Freeing an EV from a guide path, for example,
permits selective adjustment of its path to produce an EV
switch device, for example.
The present invention includes techniques for guiding
30EV's by inductance/capacitance effects, which are also
utilized for generating radio frequency signals incident on
the passage of the EV. The production of visible light
accompanying propagation of an EV in a gaseous environment
is utilized to provide optical guides for the EV path to
follow.
Since an EV represents a high concentration of electric
charge, its propagation to and arrival at an anode, for
example, can be utilized to produce fast rise and fall time
~,~, . .. . .
... .
,, . . ~:
,. :: . ,
:.! `'~' ' ' ' ' '
~ 1330827 ~
-6-
j pul~es. Such fa~t pul~e~ have variou~ application~,
including the production of an appropriate potential pul~e
on a cathode to produce pure field emi~lon productlon of
EV'~. A planar cathode generator 1~ also provided for pure
fleld emi~3ion production of EV'~. Impact of EV'~ on an
approprlate target may also be utilized to produce X-rays
from a concentrated region of the target.
The emi~lon of electron~ lncldent upon the propagatlon
of EV's may be utllized to produce controlled emi~ion of
high density electrons for variou3 applications.
Addltionally, an EV o~cillo~cope i~ dlsclosed whereby ~lgnal
analy~i~ may be effected utllizlng a deflector fleld to
affect the propagatlon of an EV whereby the lncldent
electron emi3~ion may be observed on a pho~phor ~creen, or
the like, to study the applied time-varying field, for
example. Further, an electron camera i~ provided for
ob~erving the behavior of EV's, in applied deflected fields
or otherwi~e.
The pre~ent invention thu~ provide~ the EV's
themselves, a~ well a~ variou~ technique~ for the
generation, i~olation, manipulation and exploitation of
EV's.
~ 133~827 ~
Brief Description of the Drawin~s
Fig. 1 is a top, plan view of an EV generator including
a witne~s plate for detecting the production of EV's;
Fig. 2 i~ a side elevation of the EV generator of Figo
1;
Fig. 3 is a side elevation in cros~ ~ection, partly
schematic, of another form of EV generator;
Fig. 4 i~ an enlarged side elevation in cros~ ~ection
of a wetted metal cathode for u~e in the EV generator of
Fig. 3, for example;
Fig. 5 is a view similar to Fig. 4 of another form of
wetted metal cathode;
Fig. 6 is a view similar to Figs. 4 and 5 of still
another form of wetted metal cathode;
lS Fig. 7 is a side elevation of a cathode and an anode on
a dielectric substrate;
Fig. 8 is a ~ide elevation in partial section of a
cylindrically-symmetric EV generator utilizing a separator;
Fig. 9 is a side elevation in partial ection of a
planar EV generator with a separator;
Fig. 10 is a top plan view of the separator cover shown
in Fig. 9;
Fig. 11 is a top plan view of a planar RC EV guide;
Fig. 12 is an end elevation of the EV guide of Fig. 11,
equipped with a cover;
Fig. 13 is a top plan view of another form of planar RC
EV guide;
Fig. 14 is an end elevation of the EV guide of Fig. 13;
Fig. 15 is a side elevation in cros~ section of a
cylindrically-Rymmetric RC EV guide;
Fig. 16 i~ a side elevation ~n cross section of another
form of cylindrically-symmetric RC EV guide;
Fig. 17 is a side elevation of an EV generator in
con~unction with an EV guide utilizing a gas environment;
Fig. 18 is an end elevation of the generator and guide
of Fig. 17;
Fig. 19 i3 a top plan view of an EV guide system using
optical reflectors;
. -~
, . . . . .
., , : :
~ 1330827~
^ -8
Fig. 20 is an exploded view in perspective of an LC EV
guide;
Fig. 21 is an exploded view in perspective of another
form of LC EV guide;
Fig. 22 i~ a top plan view of still another form of EV
generator in which the cathode i~ integral with a
propagating qurface for the EV'~ within a guide channel;
Fig. 23 i3 a vertical cro~q section of the EV generator
of Fig. 22 taken along section line~ 23-23 of Fig. 22;
Fig. 24 i~ an end elevation of the EV generator ~hown
in Fig~. 22 and 23, equipped with a cover;
Fig. 25 i~ a side elevation in cro~s ~ection of a
cylindrically-symmetric EV generator-launcher;
Fig. 26 is a side elevation in partial ~ection of a
lS cylindrically qymmetric EV ~elector and a guide;
Fig. 27 i~ a top plan view of a planar EV selector;
Fig. 28 is an end elevation of the EV selectGr of Fig.
27;
Fig. 29 is a top plan view of an EV ~plitter;
Fig. 30 is an end elevation of the EV splitter of Fig.
29;
Fig. 31 is a top plan view of another EV splitter;
Fig. 32 is an end elevation of the EV splitter of Fig.
31, equipped with a cover;
Fig. 33 is a top plan view of a variable time delay EV
splitter;
Fig. 34 is a fragmentary vertical cro~s section of a
portion of the ~plitter of Fig. 33, taken along line 34-34
of Fig. 33;
Fig. 35 i~ a top plan view of another form of variable
time delay EV splitter;
Fig. 36 i3 a top plan view of an EV deflection ~witch;
Fig. 37 is a vertical cross section of the EV
deflection switch of Fig. 36, taken along line 37-37 of Fig.
36;
Fig. 38 is an end elevation of the deflection switch of
Figs. 36 and 37;
Fig. 39 is a top plan view of an EV oqcilloscope;
.~.. ,, .: , , ,
, ~ 1330827
g
`:~
Fig. 40 is an end elevation of the EV oscilloscope of
Fig. 39, equipped with a cover and illu~trating the use of
an optical magnification device with the oscilloscope;
Fig. 41 i~ a side elevation, partially cut away, of a
`~ 5 electron camera ~howing an EV qource po~itioned in front thereof;
Fig. 42 is a vertical cross section of the electron
camera of Fig. 41, taken along line 42-42 oP Fig. 41;
Fig. 43 i~ a side elevation of a camera as ~hown in
Fig~. 41 and 42, mounted to view an EV o~cilloscope, and
~howing the lens sy~tem of a television camera mounted to
view the output of the electron camera;
Fig. 44 i~ a schematic representation ~howing the use
of multiple electron camera~ to observe the behavior of
lS EV's;
Fig. 45 i~ a schematic, isometric repre~entation of a
planar multielectrode EV generator;
Fig. 46 is a top plan view of another planar
multielectrode generator;
Fig. 47 i~ a vertical cro~q section of the
multielectrode EV generator of Fig. 46, taken along line 47-
47 of Fig. 46;
Fig. 48 is an end view of the multielectrode generator
of Fig~. 46 and 47;
Fig. 49 is a side elevation in cross section of an
"electrodeless" EV source;
Fig. 50 is a side elevation, partly schematic, of a
traveling wave tube utilizing EV's;
Fig. 51 is a top plan view, partly qchematic, of a
planar traveling wave circuit utilizing EV's;
Fig. 52 is a vertical cross section of a pul~e
generator utilizing EV's;
Fig. 53 is an end view of the pul~e generator of Fig.
52;
Fig. 54 is a side elevation in partial 3ection of a
field emis~ion EV generator utilizing the principles of the
pulse generator of Figs. 5~' and 53,
Fig. 55 iq a top plan view of a planar field emis~ion
~.,,, ~, .
.. . .
,. .- -~ ':, :-
" i330827 ~
-..... . -10-
: , ,
. . . .
.: '
EV generator;
Fig. 56 is a circuit diagram for operating the field
emi3sion EV generator of Fig. 55;
Fig. 57 is a side elevation in partial sectlon of an X-
ray generator ut$1izing EV's;
Fig. 58 is an exploded, i~ometric view Or a gated
electron ~ource utilizing EV's;
Fig. 59 is an exploded, isometric view of an RF ~ource
utilizing EV'q;
Fig. 60 i9 a schematic, pictorial view of an EV; and
Fig. 61 is a qchematic, pictorial view of a chain of
EV's.
,
,~,
,
. 5
,:i i
~, r'~
!~ 1 3 3 0 8 2
. ^ ~
', -1 1-
i Descri~tion of Preferred Embodiment~
! 1. Definition and Some EV Properties
An Ev is a discrete, self-contained, negatively charged
bundle of eleotrons. While not yet fully under~tanding the
configuration of an EV, I believe the self-containment to be
due to electromagnetic fields set up between the electrons
within the bundle, based upon my many ob~ervations of EV
behavior. This, of course, is in sharp contrast to a
conventional electron beam in which the containment of
electrons is due either to an external electrostatic field
or an external magnetic field. As is well known in the art,
electrons, each being negatively charged, tend to repel each
other. -
It should also be appreciated that even though the EV
is a self-contained bundle of electrorls, it does prefer to
communicate with other objects or entities, such a~ other
EV'~, dielectrics and electrodes, for example, as contrasted
with going off on its own, and tends to come apart after
some period of time if there is nothing with which to
communicate.
Primary characteristics of an EV include its relatively
small ~ize (for example on the order of one micrometer in
lateral dimension, but can be larger or as small as
0.1 micrometer), and high, uncompensated electron charge
(that is, without positive ion~, or at least with an upper
limit of one ion per 100,000 electron chargesj, typically on
the order of 1011 electron charges. The minimum charge
observed for a one micrometer EV is 108 electron charges.
The charge density of an EV approximates the average density
of`a solid, that is, on the order of 6.6 x 1023 electron
charges/cm3, but without being space charge neutralized by
ions or having relativistic electron motion. The velocity
attained by an EV under applied fields (on the order of one
tenth the speed of light) indicates that the EV charge-to-
mass ratio is ~imilar to that of an electron, and deflection
of EV's by fields of known polarity shows that EV'~ respond
as electron~, that is, as negatively charged entitie~.
,- ~ ,: ,- . . ~ ,
.~: ~ ~ ~ , . - , .~.:, .
1330827
As best a~ can be determined at p~esent, the ~hape of
an EV i~ mo~t likely generally ~pherical, but may be toric,
and could have fine structure. As ~chematically illu~trated
in Fig. 60, an EV i~ illu~trated a~ having a central ~phere
800 of ~elf-contained electrons, ~urrounded by an
electromagnetic field 801. Coupling between EV'~ produce~
quasi ~table structure~. However, lone EV'~ are rarely
observed. EV'~ exhibit a tendency to link up like bead~ in
a chain, for example, a~ ~chematically illustrated in Fig.
61, wherein the EV bead~ in the chain may be somewhat free
to rotate or twi~t about each other under the influence of
external force~ or internal force~. The chain~, which are
clo~ed, may be ob~erved to form ring-like ~tructureq a~
large as 20 micrometer~ in diameter, and multiple chains may
al~o unite and mutually align in relatively orderly
fa~hion. In the chain 810 of Fig. 61, the ten EV'~ 812,
814, 816, 818, 820, 822, 824, 826, 828 and 830 are ~hown
generally in a circular pattern. Spacing of EV beads in a
chain i~ normally approximately equal to the diameter of the
individual bead~. Spacing of one chain ring ~rom another i~ ~
on the order of one ring qiameter. A one micrometer wide -
ring of ten EV beads, which i~ the typical number of beads
in a chain, may include 1012 electron charges. Individual
EV bead~ may be ob~erved within a chain ring. An EV entity,
which i~ in the nature of a non-neutral electron pla~ma, i~
most ~trongly bound, with the binding force between EV bead~
in a chain being weaker, and finally the binding between
chain~ of bead~ being the weake~t. However, all of the
binding energie~ appear to be greater than chemical binding -~
energy of material~. Additional EV propertie~ are di~cu~ed
hereinafter.
: .
~,
~ - . . .
/~
3~3~27 1
-13-
2. Generators
An EV may be generated at the end of an electrode that
has a sufficiently large negative voltage applied to it.
Figs. 1 and 2 illu~trate an EV generator, shown generally at
10, including a cathode 12 generally in the form of an
elongate rod having a neck portion 12a ending in a point and
directed generally downwardly toward an anode plate 14
separated from the cathode by an intervening dielectric
plate 16. A~ indicated in the drawing, the anode, or
collector electrode, 14 is maintained at a relatively
positive voltage value, ~hich may be ground, and a negative
pul~e on the order of 10 kv is applied to the cathode 12 to
generate an intense electric field at the point of the
cathode. With the resulting field emission at the cathode
tip, one or more EV's are formed, generally in the vicinity
of where the point of the cathode approaches or contacts the
dielectric at A. The EV's are attracted to the anode 14,
and travel across the surface of the dielectric 16 toward
the anode, generally along a path indicated by the dashed
line B, for example, as long as the dielectric surface is
uncharged. Propagation of one, or several EVS, along the
dielectric surface may leave the surface locally charged. A
subsequent EV will follow an erratic path on the surface
unless the surface charge is first dispersed, as discussed
in detail hereinafter. The insulating dielectric plate 16,
which is preferably of a high quality dielectric, such as
quartz, prevents a direct discharge between the cathode 12
and the anode 14, and also serves to provide a surface along
which the EV's may travel.
If desired, a witness plate 18 may be positioned
adjacent the anode 14 to intercept the EV's from the cathode
12. The witness plate 18 may be in the form of a conducting
foil which will sustain visible damage upon impact by an
EV. Thus, the witness plate 18 may be utilized to detect
the generation of EV's as well as to locate their points of
impact at the anode 14. Additionally, an EV propagating
across the dielectric ~urface will make an optically vi~ible
streak on the surface. ~s discussed in further detail
!
1330827
~ -14-
hereinafter, other component~ may be utilized in conjunction
with the generator lO to further manipulate and/or exploit
the EV's thus generated.
The generator lO may be located within an appropriate
enclosure (not ~hown) and thu~ operated in vacuum or in a
controlled gaseous atmo~phere as de~ired. In general, all
of the components disclosed herein may be so positioned
within appropriate enclosures to permit selection of the
atmosphere in which the componentq are operated. Terminals
or the like, and gas tran~mi~sion line~ may be utilized to
communicate electrical ~ignals and selected gas at de~ired
pressure through the enclo~ure walls.
The scale indication of lO mm included in Fig. l iq a
typical dimension for EV generating components. Generally,
when EV's are generated and manipulated in small numbers,
they can be made and guided by small structures. Even when
large structures are used, an EV seeks the ~mallest details
of the gross structures and i~ guided by them and interacts
most actively with them, leaving the larger details
unattended. To a first approximation, generation and
manipulation of individual EV bead~ may be accompli~hed with
structures having overall dimensions of as little as ten
micrometers.
Generally, very stable materials are de~ired for use in
the construction of structures to generate, manipulate and
exploit EV's, including refractory metals and dielectrics
cho~en to approach a~ closely as possible the binding energy
of an EV, so as to preserve the life of the structures.
Some dielectric materials, such as low melting point
plastic, are not as preferably as other materials, for
example, such as ceramic.
With any type of EV generator, and whether dc or a
pulse signal is applied to the cathode, it is neces~ary to
complete the current flow path around a loop by using an
electrode of some type to collect the EV (except in the case
of "electrodeless" source~ as discu~sed hereinafter).
Another form of EV generator is shown generally at 20
in Fig. 3, and include~ a cylindrically ~ymmetric cathode 22
,.,
'`'~
.
~330827
having a conical end facing but displaced from an
anode/collector electrode 24 which is also cylindrically
symmetric. An operating circuit includes a load resistor 26
connecting the anode 24 to ground, while a current limiting
5input resistor 28 is interpo~ed between the cathode 22 and
an input terminal 30. The anode 24 is equipped with an
output terminal 32 to which may be connected ancillary
equipment. For example, detection equipment (not shown),
such as an oscillo~cope, may be ~oined to the system by
10terminal 32 whereby the impact of E~'s on the anode may be
noted.
An enclo~ure, such a~ within a cylindrical glas~ tube
34, may be provided whereby the environment in the gap
between the cathode 22 and the anode 24 may be controlled,
15and maintained either in vacuum or at a selected gas
pressure. The tubing 34 may be appropriately sealed and
fitted with communication lines (not shown) to a vacuum pump
and/or gas supply to control the environment within the
tube.
20The cathode 22 may be driven by a negative-going pulse,
or a direct current, of approximately 2 kv relative to the
anode. The length of the negative pulse may be varied from
a few nanoseconds to dc without greatly influencing the
production of EV's. Under long pulse length conditions, the
25input resistor 28 must be chosen to prevent a sustained glow
discharge within the glass tube. Under high vacuum condi-
tions, or low pressure such as 10-3 torr, the discharge is
easily quenched and the resistor 28 may be eliminated, but
for a gaseous environment of higher pressure, a value of the
30resistor mu~t be chosen that i9 consistent with the gas
pressure used so as to quench the discharge. For operation
in both a vacuum and gaseous regime using a pulse length of
0.1 microsecond, for example, a typical resistor value of
500 to 1500 ohms can be used.
35In high vacuum operation of the generator 20, the
~pacing between the cathode 22 and the anode 24 should
preferably be less than 1 mm for a 2 kv signal applied to
the cathode. For operation in gases at pressures of a few
.. . .
~`
~ 1330827 ~
-16-
torr, the distance between the cathode 22 and the anode 24
may be increased to over 60 cm provided a ground plane 36 is
used ad~acent the glass tubing as shown. The ground plane
36 may extend partly around the tubing 34, or even
circumscribe the tube. For particular applications, the
glass tube 34 can be replaced by other structures to guide
EV's, as discussed hereinafter, and various circuits can be
devised to take advantage of various EV propertie~.
r
~;
-- 1331~827
3. Cathode~
The cathode~, such as 12 and 22 discus~ed hereinbefore,
may be pointed by any appropriate technique, such a~
grinding and polishing. and even chemical etching. to
achieve a sufficiently sharp point to allow the
concentration of a very high field at the end of the
cathode. Under normal conditions, as EV's are generated at
the tip of such a metallic electrode, the electrode material
is dispersed and the cathode point or other configuration is
destroyed by the energy dis~ipated in it, and the voltage
required to produce EV's increa~es. However, the cathode
may be coupled to a source of liquid conductor, and the tip
of the electrode regenerated in a very short time. Fig. 4
show~ a metallic electrode 40 that is wetted with a
conductive sub~tance 42 coated onto the cathode whereby the
coating material may undergo qurface migration to the
pointed tip of the electrode. The migrating material renews
the tip of the electrode to maintain a sharp point as EV
generation by the electrode tends to deteriorate the
electrode tip. Surface ten~ion of the coating material 42,
its destruction at the tip, and the electric field generated
at the cathode combine to propel the migration of the
coating substance toward the tip.
In Fig. 5 an electrode 44 i~ surrounded by a tube 46
whereby an annular ~pacing 48 is defined between the outer
surface of the electrode and the inner ~urface of the
tube. The spacing 48 serves to maintain a re~ervoir of
coating material 50 which is held within the ~pacing by
surface ten~ion, but wets the cathode and migrates to the
tip of the cathode in forming a coating 52 thereon to
maintain an appropriately sharpened cathode point. The
reservoir tube 46 is preferably a non-conductor, ~uch a~
aluminum oxide ceramic, to prevent unwanted electron
emission from the tube as well as unwanted migration of the
wetting material along the tube. Otherwise, a conductor
tube may be used aQ long as it is not too cloQe to the
cathode tip, whereupon the tube may emit electrons. The
coating material 50 may. in general. be any metallic liquid
~, . . ~ . . ~ ; .:
': 133~27
-18-
such as mercury, which may appropriately migrate over an
electrode 44 constructed of copper, for example.
The cathodes 40 and 44 of Figs. 4 and 5, re~pectively,
are designed for EV emi~sion from a specific point. In Fig.
6 a tubular cathode 54 features a conically shaped interior
at one end forming a sharp, circular edge, or line, 56 at
which EV's are generated. The cylindrical portion of the
interior of the line cathode 54 defines, by means of surface
tension, a reservoir of coating material 58 which wets and
migrates along the conical interior surface of the cathode
toward the emitting edge 56. Thus, the migrating material
5~ renews the circular edge 56 to keep it appropriately
sharp for EV generation.
Generally, for a source that can be fired repeatedly to
produce EV's, a migratory conductor is needed on a
conductive substrate that has a field-enhancing shape. The
sharpened point of a cathode, such as shown in Fig. 4 or 5,
may become further sharpened by the effect of the metallic
coating wetted thereon being drawn into a microscopic cone
by the applied field. Similarly, the coating material in a
tubular cathode, such as shown in Fig. 6, is drawn to the
circular edge due to field effects to provide a particularly
sharp edge including microscopic emitting cones.
A wide variety of materials can be used to construct
wetted cathodes in general. Typically, for room temperature
operation of an EV generator, the cathode may be constructed
of pointed copper wire coated with mercury. Alternatively,
mercury can be coated onto silver or molybdenum. Similarly,
gallium indium alloys or tin lead alloys can be used to coat
a vaFiety of substrate metals to form cathodes. Examples of
cathode structures for use at high temperatures include
aluminum coated titanium carbide for operations at 600C,
and boron oxide glass coated tungsten in operations at
approximately gaooc.
Non-metal conductive coatings may also be used. For
example, coatings of glycerin doped with potassium iodide or
sodium iodide, and nitroglycerin doped with nitric acid,
have been ~uccessfully used with a variety of metallic
r
". ''
~ h 7
(- ~3~0827 ~
_19_
sub~trates such as copper, nickel, tungsten and
( molybdenum. The glycerin is nitrated by including acidt or
doped, to impart some conductivity to the organic
material. However, it is not neces~ary to dope for
conductivity if the coating material is kept to a very thin
layer. Polarization of such material i8 sufficient to allow
the material to be moved in a field to thus pump the
material to a field enhancing tip. -
It will be appreciated that operation of a wetted
source, particularly in a reduced ambient pressure
environment, even a vacuum, is accompanied by the wetting
material vaporizing, or yielding ga~eous product~. Thus,
the metal-wetting material forms a vapor. Organic or
inorganic gases may be acquired depending on the wetting
substance. Field emission is accompanied by current through
the cathode which heats the cathode, causing the
vaporization of the wetting material. Field emitted
electrons impact and ionize the vapor particles. The
resulting positive ion cloud further enhances field emission
to produce an explosive-like runaway process resulting in a
high, local electron density.
Variations of wetted cathodes may enhance migration of
wetting material, return evaporated material to the source,
keep the field producing structure sharp and/or help reduce
ioni~ation time to allow high pulsing frequencies to produce
EV's. To take advantage of the regeneration provided by
wetting cathodes, the pulse rate of the signal applied to
the cathode to generate EV's must be low enough to allow
migration of the coating material to restore the point or
line between pulses. However, for extended, or line,
source~, such as the circular cathode 54 of Fig. 6, the
pulse rate may be raised to much higher values than is
practical for use with point sources since the complete
regeneration of the line between pulse~ by coating migration
3~ is not necessary. Some portion of the line cathode is
generally left sharp for subsequent EV production after
production of EV's elsewhere along the line.
Fig. 7 qhows an EV ge!nerator 60 including a ceramic
: :
r,: .
133~827
-20-
base 62 having a planar, or surface, cathode 64 positioned
along one surface of the base, and a planar anode, or
counterelectrode, 66 positioned along another surface of the
base generally opposite to the position of the cathode. The
cathode 64, which is effectively another form of extended or
line source, may be coated with a metallic hydride, such as
zirconium hydride or titanium hydride, to produce EV's.
Such a cathode continues effective provided hydrogen is
recharged into the hydride. This can be done by operating
the generator, or source, in a hydrogen atmosphere so that
the cathode is operating in the thyratron mode, which is a
known hydride regeneration technique. However, since there
is no flow of wetting material onto the cathode base
material, after a period of use the coating material
disperses and the source fails to fire. Consequently, in
general. the surface source 64 has a shorter effective life
than cathodes on which migratory material is deposited, such
as those shown in Figs. 4-6. Additional details of the
construction and operation of a surface generator such as
illustrated in Fig. 7 are provided hereinafter.
.. - .
-` 12330827
4. Separators
In general, the production of EV'~ is accompanied by
the formation of a plasma discharge, including ions and
dlsorganized electrons, generally where the EV's are
produced at the cathode, wlth the plasma charge density
being at least 106 electron charges per cubic micrometer,
and typically 108 charges per cubic micrometer. In the case
of a relatively short distance between cathode and anode of
a source, the high plasma density accompanying the formation
of the EV's is usually produced in the form of a local
spark. As the distance between the cathode and the anode is
increased, EV production and transmission is also
accompanied by the formation of streamers, that is, excited
ions in a gaseous mode along the path of an EV which yield
light upon electron transition. As noted hereinbefore, an
EV itself comprises an extremely high total charge
density. Typically, a chain ring of ten EV beads, with each
bead approximately 1 micrometer in width, may contain 1012
electron charges and, moving at approximately one-tenth the
speed of light, may pas~ a point in 10-14 seconds,
establishing a high current density easily distinguishable
from ordinary electron current. Generally, in the case of a
pulsed source, an EV may be expected to be formed for each
pulse applied to the cathode, in addition to the extraneous
charge production that may accompany EV production.
The various components of the plasma discharge present
when EV's are formed are considered as contaminants to the
EV, and are preferably stripped away from the EV
propagation. Such stripping can be accomplished by
enclosing the EV source in a separator, positioning an
aperture or small guide groove between the source and the
extractor electrode, or anode. A counterelectrode i9
provided on the enclosure for use in the formation of the
EV's. The discharge contaminants are contained within the
separator while the EV's may exit through the aperture or
groove toward an extractor electrode.
An EV generator shown generally at 70 in Fig. 8,
includes a cylindrically-symmetric and pointed cathode 72,
~ 1330827 ~
-22-
which may be mercury wetted copper, for example, aDd a plate
anode 74, and is equipped with a cylindrically-symmetric
separator 76. The separator 76 includes a generally tubular
member, constructed preferably of a dielectric, for example
a ceramic such as aluminum oxide, that tapers beyond the
point of the cathode 72 in a region 78 including a
frustoconical exterior surface and a frustoconical interior
surface of smaller angle of taper to form an aperture 80
defined by a relatively sharp circular end of the tubular
member. When a dielectric is used for the tunnel 76, a
counterelectrode 82 is formed on the exterior of the tunnel
and maintained at a positive potential relative to the
cathode 72, while the anode 74 is positive relative to the
counterelectrode. Typically, the voltage values may be in
the range of 4 kv, 2 kv and zero on the extractor anode 74,
the counterelectrode 82 and the cathode 72, respectively.
The electrode 82 not only provides the relative positive
potential for the formation of the EV's but acts as a
counterelectrode for propagating the EV's through the nozzle
aperture 80, while the di~placed anode 74 represents a load,
for example, and may be replaced by any other type of
exploiting load. Other materials, such as semiconductors,
may be used to form the tunnel 76 with appropriate
electrical isolation from the cathode 72. In Ruch cases,
2~ the tunnel material itself can serve as a counterelectrode.
Since an EV induces an image charge in a dielectric
separator 76, the EV tends to be attracted to the dielectric
~urface. However, the various contaminants of the formation
discharge, including electrons and ions, may be repelled by
the tunnel separator 76, at the same time the EV's are
attracted to the tunnel. Thus, the EV's may emerge through
the aperture 80 free of the discharge contaminants, which
are retained within the separator 76. The cross section of
the aperture 80 must be ~uch as to allow emergence of EV's
while at the same time providing a sufficiently narrow
channel to retain the di~charge contaminants and prevent
their passage through the aperture.
The construction of the generator 70 with the tubular
~'~
;~ ~33G827
separator 76 having a small aperture 80 is relatively
convenient for u~e with various environments between the
cathode 72 and the anode 74. For example, the exit ~ide of
the nozzle formed by the separator 76 with the aperture 80
may be sub~ect to vacuum or selected gas pressure as
desired. The formation side of the nozzle, that is, the
interior of the separator 76 in which the cathode 72 is
positioned, may be vented to either vacuum or a gaseous
region as selected, different from the exit side
environment. Appropriate pumping can be utilized to
maintain the desired environments.
While the separator 76 illu~trated and described
hereinabove is shaped like a funnel, I have found that a
square box (not shown) having a small aperture, similar to
aperture 80, for the EV's to exit, works quite well in
separating the EV's from the remainder of the electrical
discharge, which as stated before, may include electrons,
positive and negative ions, neutral particles and photons.
Fig. 9 shows an EV generator, indicated generally at
84r equipped with a separator designed for use in a planar
construction for an EV generator. A dielectric base 86 is
fitted with a surface cathode 88. A separator in the form
of a dielectric cover 90 extends over and beyond the cathode
88, and terminates in a sloped exterior surface which,
coupled with a sloped interior surface of smaller angle of
slope, provides a relatively sharp edge suspended a ~hort
distance 92 above the surface of the base 86. As
illustrated ir. Fig. 10, the separator 90 is also pointed in
the transverse direction at the edge toward the spacing 92,
and features walls 94 which cooperate with the sloped
interior surface to define the peripheral limits of the
region effectively enclosed between the separator cover and
the base 86. The outer flat surface of the cover 90 is
partially coated with a counterelectrode 96, which extends
downwardly approximately two-thirds the length of the sloped
outer surface of the cover to provide a relative positive
potential for the formation and propagation of EV'~ from the
cathode 80. A target anode 98 is positioned on the opposite
, . .
. ,~, ~
:
.
~,'~: - ... . .
:, .
~ 3 3 0 8 2 7 !
-24-
side of the ceramic base 86 to collect propagated EV's, and
( may be replaced by some other load used in manipulating
and/or exploiting the generated EV's.
The separator 90 functions essentially like the
~eparator 76 of Fig. 8 in that the EV's generated by the
cathode 88 in Fig. 9 are attracted forward by the
counterelectrode 96 of the cover 90 toward the opening 92,
while extraneous discharge contaminants are retained within
the cover 96. Alternatively, the cathode 88 may be set in a
groove (not shown) extending beyond the back of the cover
90, and the cover set down on the base 86. A small groove
may be provided on the underside of the cover, or on the
base, ln the area 92 to allow passage of EV's out of the
cover enclosure. The groove of the cathode 88 may continue
through the area 92 to allow exit of the EV's from under the
cover 90. Additionally, the counterelectrode 96 may be
deleted if the anode 98 extends to the left, as seen in Fig.
9, to underlie the area 92.
The base 86 and the separator cover 90 may be
constructed from ceramic materials such as aluminum oxide,
and the counterelectrode 96 and the anode 98 may be formed
from a conductive layer of silver fired onto the ceramic
substrate, for example. The cathode 88 may be formed of
silver fired onto the dielectric, and wetted with mercury,
for example.
Other coating processes for constructing conductor
patterns, such as thermal evaporation or sputtering, may be
used to form the counterelectrodes of the two separators 76
and 90 shown in Figs. 8 and 9, respectively. The openings
provided by the separators must be sufficiently small to
permit emergence of the EV's while stripping away the
discharge contaminants. For example, the aperture 80 of the
separator 76 in Fig. 8 may be approximately 0.05 mm in
diameter for the generator operating at 2 kv, and with a
circular lip khickness of approximately 0.025 cm. The lip
and opening sizes provided by the cover separator 90 of Fig.
9 may be comparable. In either case, smaller openings can
tolerate smaller voltage~ and still filter contaminants
: :
1 33 ~8~7 ~
. -25-
effectively. Generally, the exact cro~-sectional ~hape of
the separator is not of primary importance for the filtering
function.
-, -:
~'i
263~os27
5. RC Guide~
In general, an anode cooperate~ with a cathode in the
application of appropriate electrical potential to generate
EV's, and may serve as the target or load of the generator,
and actually be impacted by EV's. In general, a
counterelectrode is not impacted by EV's, but i~ u~ed in the
manipulation and control of EV's, and may be used in the
generation of EV's. For example, the counterelectrode~ 82
and 96 of Figs. 8 and 9. respectively, contribute to drawing
the EV's forward away from the region of EV generation at
the respective cathodes, but the EV's continue on to
po~sibly ~trike the anodes 74 and 98, respectively, although
both counterelectrodes 82 and 96 also prov~de the EV
formation voltage. As discus~ed more fully hereinafter, an
EV may move along or close to the surface of a dielectric
material placed in the path of propagation of the EV. If a
ground plane, or counterelectrode, at an appropriate
positive potential, relative to the generating cathode, is
positioned on the opposite side of the dielectric material,
the EV propagating on the cathode side of the dielectric
material will tend to be attracted to the counterelectrode
through the dielectric, and this attraction may be used to
influence the path of the EV along the dielectric as
discussed more fully hereinafter; particularly in the case
of RC (re~i~tance/capacitance) guides for EV's.
If an EV is directed toward a dielectric ~tructure,
backed by a counterelectrode or anode at relative positive
potential, the EV may move on the surface of the dielectric
in an apparent random fashion. However, the path of the EV
is determined by local electrical effects, ~uch as the
dielectric polarizability, surface charge, surface
topography, thickne~ of the dielectric and the initial
potential of the backing electrode along with its
. .
conductivity. The ma~or ~echanism that affects the movement
of EV's on dielectric ~urfaces i~ the polarizability of the
dielectric producing an image force that attracts the EV to
the dielectric, but doesn't; move the EV forward. Even in
the ab~ence of a counterelectrode at an appropriate
',i- ',' : : '' ' ' ,: : . , ' . ' ,
~ 3 3 0 8 2 7!
27-
potential, the induced image charge tends to attract an EV
to the dielectric surface. The EV cannot go lnto the
dielectric. Consequently, an EV will tend to move across
the surface of a dielectric and, when an edge or corner of
the dielectric material is reached, the EV will, in general,
go around that corner. As noted hereinbefore, EV's tend to
follow fine structural details, and this is evident from the
guiding effect caused by surface scratches and
imperfections. Generally, any intersection of two
dielectric surfaces or planes having an angle of
intersection less than 180 will tend to guide the EV along
the line of intersection.
Figs. 11 and 12 illustrate an EV guide component shown
generally at 100, including a dielectric base member 102
featuring a smooth groove 104 providing an enhanced guide
effect. A counterelectrode plate 106 covers most of the
opposite surface of the base 102 from the groove 104, and
may be maintained at relative positive potential with
respect to the emitting cathode, which is generally directed
toward one end of the groove. The guide component 100 may
be utilized, for example, in conjunction with an Ev
generator as illustrated in Figs. 1 and 2, and a separator
such as shown in Figs. 9 and 10. However, such a guide
member 100 may be utilized with virtually any EV source and
other components as well. An optional top cover 108, of
dielectric material as well, is illustrated in Fig. 11 for
placing over the groove 104, in contact with the base 102.
The width and depth of the groove 104 need only be a
few micrometers for guiding small numbers of EV's. However,
as the power to be handled increases and the number of EV's
increases, crowding may become a problem and it is necessary
to increase the size of the groove. The cross-sectional
shape of the groove 104 is not of primary importance in its
ability to guide EV's. Wit-h EV's generated by a generator
such as shown either in Figs. 1 and 2 or in Fig. 3, and
coupled to ~ guiding comlponent by a separator such as
illustrated in Figs. 8 or 9 and 10, and with the guIding oomponent,
such as shown in Figs. 10 and 11, comprising a fused silica ~;~
~33~827 --
-28-
or aluminum oxide dielectric baqe with an overall thickne~s
of 0.0254 cm and having a groove 104 of 0.05 mm in depth and
0.05 mm in width, the guiding action is demonstrable.
Figs. 13 and 14 ~how a variation of a planar guide
component, indicated generally at 110 and including a
dielectric ba~e 112 with a dielectric tile 114 po~$tioned on
and appropriately bonded to the ba,~e. The intersection of
the surface of the base 112 with the surface of the tile
meeting the base at a 90 angle of inter~ection (that i~,
one half of a groove such as 104 in Figs. 11 and 12) would
provide a 90 NV" along which EV's could propagate. The
guiding effect, however, is enhanced by a beveled edge a~
shown, ~et at approximately 45, along the tile surface
intersecting the base to form a groove indicated generally
at 116. A counterelectrode plate 118 i8 pOS itioned along
the opposite surface o~ the base 112 from the tile 114. A
collection of tile~ ~uch as 114, complete with beveled edges
to form grooves such as 116, may be po~itioned along the
base 112 in a mosaic to define an extended guide path. The
guide component 110 may be utilized with virtually any other
components u~ed to generate, manipulate and/or exploit EV's.
The guiding action on an EV may be enhanced by u~e of a
tubular dielectric guide so that the EV may move along the
interior of the tube. Fig. 15 illustrates a tubular
dielectric guide member 120 having an interior, ~mooth
passage of circular cro~s section 122 and coated on the
outside with a counterelectrode 124. The cross-sectional
area of the interior channel 122 should be slightly larger
than the EV bead or bead chain to be guided thereby for best
propagation properties.
The glas~ tube 34 with the ground plane 36 encircling
the tube, shown with the generator 20 in Fig. 3, is a guide
of the type ~hown in Fig. 15. For different applications,
the glass tube 34 in Fig. 3 may be replaced by a guide of
another type.
Fig. 16 illustrates a guide member constructed
generally as the reverse of that of Fig. 14, namely, a
dielectric tubular member 126 having an interior channel 128
~ 133~827 ~
-29-
coated with an interior counterelectrode 130, and providing
the exterior, generally cylindrical surface 132 as a guide
surface in con~unction with the dielactric structure itself
and the counterelectrode 130. In this instance, an EV may
move along the exterior surface 132, attracted to the guide
member by the image charge generated due to the presence of
the EV, and also by the effect of the counterelectrode 130
maintained at a relative po~itive potential.
In general, the dielectric guides of Fig~ 16, as
well a~ other dielectric components, can be appropriately
doped for limited conductivity to limit or control stray
charge, as di~cussed more fully hereinafter. An EV moving
within the guide ~tructure of an RC guide device provides a
temporary charge on the guide as noted hereinbefore, and
another EV will not enter the immediate high charge region
of the guide due to the first EV, but can follow after the
charge on the dielectric dissipates after pa~sage of the
first EV.
If the groove, or tunnel. used a~ a guide through or
acros~ a dielectric material is too narrow in cros~ section
compared to the size of an EV, the EV passing along the
guide may effectively cut into the guide material to widen
the path. Once a channel ha~ been bored out by an EV in
thi~ manner, no further damage i~ done to the dielectric
material by subsequent EV's propagating along the guide.
Typically, a channel of approximately 20 micrometers in
lateral dimen~ion will accommodate EV pa~sage without boring
by the EV. This i~ about the lateral dimen~ion of an EV
bead chain formed into a ring that can be produced with a
given ~ource. The guide groove can be made larger or
smaIler in cros~ ~ection to match larger or smaller EV'~
depending on the circumstances of their production.
, ~
;
~,. ~. . . .
` 1~1.`.~
.~ 1330827 ~
-30-
6. Gaseous Guides
Any of the guide structures illustrated in Figs. 11-16
may be utilized either in vacuum or in a selected gaseous
environment. However. the use of gas at low pres~ures in
guide members can produce another beneficial effect in the
manner of guiding EV's formed into a chain of beads, for
example.
In some in~tances, EV's formed from high powered
sources may be composed of beads in a chain configuration.
Such a chain group may not propagate well on a particular
solid guide surface due to the very tight coupllng of the
beads in the chain and the disruption that surface
irregularities caused in the propagation of the
configuration. In a low pressure gas atmosphere, typically
in the range starting at about 10-3 torr and extending
through 10-2 torr, the EV chain is lifted a relatively short
distance from the dielectric surface and no longer interacts
in a disruptive fashion with the surface, with the result
that transmission efficiency i~ increased. Then, in
general, for a given applied voltage, EV's can be formed
with greater separation between cathode and generating
anode, and can traverse greater distances between
electrodes. Evidence from witness plates appears to
indicate that, moving relatively free of a solid surface, a
bead chain tends to unravel and propagage generally as a
circular ring, lying in a plane perpendicular to the
direction of propagation. In general, as the gas pressure
is increased, the EV may be lifted further from the solid
surface. For gas pre~sures above a few torr, EV's in
general move off of the solid surface entirely, and the flat
solid surface no longer functions as a guide. However, a
guiding effect may still be realized with ~uch higher gas
pressure for EV's moving along the interior of a closed
guide, such as that illustrated in Fig. 15.
Although a wide variety of gases appear to be useful to
produce the lifting effect on EV's and EV configurations,
the high atomic number gases such a~ xenon and mercury
perform particularly well. The enhanced guiding action on
;~
.~ . ....
'~, ,
~ ~ 33~827 ~
-31-
such EV configurations and single EV's works well on the
inside of dielectric guide enclosures such as those
illustrated in Figs. 11-15, and also works well on single
plane ~urfaces.
Figs. 17 and 18 illu~trate a guide device constructed
to utilize a "cushion" of gas to maintain EV's lifted from
the guiding surface~ while yet providing a groove, or
trough-like guiding structure. The "gas" guide, shown
generally at 136, includes a trough formed from a dielectric
block 138, which may, for example, be in the form of a glaze
coated, porous ceramic. The dielectric block 138 features a
counterelectrode 140 on the bottom of the block, and further
has coatings of resistor material 142, described hereinafter
in the section entitled "Surface Charge Suppression," along
the interior lower portions of the trough, or groove, to
resist movement of EV's along the ~o-coated surface out of
the trough provided by the block 138. The guide component
136 is connected to a gas communicating line 144 by means of
a fitting 146, and which feature~ an internal passage 148
through which gas selectively communicated to the guide may
pass to the bottom of the block 138 from a source (not
shown). The bottom of the dielectric block 138 is not
glazed at the intersection with the fitting passage 148 so
that gas may enter the porous interior of the block. The
~5 glaze coating and the resistor material coating 142 are
scratched, or cut, along the bottom of the V-shaped trough
to permit gas to emerge from the interior of the dielectric
block 138. The entire arrangement is enclosed for selective
control of the environment, and a vacuum pump system is
applied to the enclosure to pump away the gas emerging
through the block 138. Thus, gas introduced into the porous
block 138 through the fitting 146 emerges along the bottom
of the trough, and, in dispersing upwardly throughout the
trough, provides a gas pressure gradient. The concentration
of the gas thus varies from heavy to light going from the
bottom of the trough upwardly. A pointed cathode 150, such
as a mercury-wetted copper wire, extends downwardly toward
the bottom of the trough at a short distance from the
.
~yy `
~330827 ~
-32-
beginning of the resistor coating 142, and may be maintained
with the cathode terminal point a short distance above the
dlelectric material of the trough.
In operation, a negative pulse signal of about 2 kv (or
higher if the cathode tip is not sufficiently ~harp) may be
applied to the cathode 150 while the counterelectrode 140 i8
maintained at ground potential, that is, relatively
positive, to generate EV's at the tip of the cathode well
within the depth of the trough formed by the dielectric
block 138, where the ga~ pre~sure is highest. The EV's
propagate along the length of the trough a~ ~elected gas is
introduced into the trough through the communication line
144, and the EV's lift o~f in the gas layer just above the
bottom of the trough, still attracted to the dielectric
block 138 by the image charge, or force, of the dielectric
material and the potential of the counterelectrode 140. The
wedge-shaped gas pressure gradient provided by the trough
contains, or ~Ifocuses~ the ga3 cushion effect to help keep
the EV's within the confines of the trough. However, a
~ufficient gradient would be provided even if the trough
were replaced with a flat ~urface having a similar cut in
the glaze coating and the re~i~tor material coating 142 so
that, and further in view of the image force effect and
counterelectrode potential, EV's would be guided along the
dielectric block, just generally above the cuts in the
coatings. Further, from the foregoing discu~sions
concerning the effect of low gas pre~sure on EV propagation
over dielectric surfaces, it will be apprepciated that EV's
will lift over such a guide surface with no gradient present
in the gas pressure.
~---
.;
1330827 ~
~ 33-
7. Optical Guides
An EV moving through a purely, low pressure, gaseous
pha3e where no RC guiding structures are pre~ent, is
accompanied by the formation of a visible ~treamer. A
narrow beam of light appear~ to precede the ~treamer, and
may be due to ionization of the gas by the streamer. In any
event, the EV follow~ the path defined by the streamer, and
the streamer appears to follow the propagation of the
light. Such an effect also occur~, for example, when EV's
move over a guide surface in a gaseous environment, such as
an environment of xenon gas. When an EV is propagated on or
along the surface, it travels in a straight line if the
surface is very clean. (Surface charge effects di~sipate
after an EV is propagated in a gas environment.) The
forward-looking light from the streamer defines a straight
path followed by the streamer and therefore, the EV. If
thiq light path is deflected by objects on the ~urface, the
streamer will deflect, and the EV wilI follow the new
path. Only a small disturbance is needed to start the
change in path. Once the path i~ described, it will remain
for future use as long as the ~treamer persists.
Fig. 19 illustrates an optical guide for use in a
ga~eous environment. A dielectric plate 152 has a path 154
schematically noted thereon, proceeding from left to right
as viewed in Fig. 19. The path 154 may be a scratch on the
surface of the plate 152 or an actual guide groove in the
plate. A counterelectrode (not visible), at an appropriate
potential, may be po~itioned on the underside of the
dielectric material I52 to aid in the propagation of EV's
over the dielectric surface. A reflecting surface 156 is
po~itioned to intersect the EV path along the dielectric
plate 152, indicated by a dashed line. The surface 156
reflects the light incident thereon, apparently according to
the laws of optics, with the result that the EV path is
likewise deflected as indicated. A second reflecting
surface 158 intersects the new, deflected light path, and
deflects the path to a new direction. Consequently, an EV
will trace the light path, indicated by the dashed line,
~''' ' :
~`
.
~330827 ~
-34-
guided by both reflectors.
Each of the optically reflecting device~ 156 and 158 is
preferably a front ~urface reflector of high dielectric
constant mater~al with good reflection in the ultraviolet
region. The angle of reflection determines the eventual EV
path in each case. The change in direction of the light
path effects a change in direction of the streamer, and the
EV followq the streamer along the path defined by the
light. A gas presqure of several torr can be utilized above
the dielectric curface where the EV's propagate and are
appropriately guided. The reflectors 156 and 158 need only
be a fraction of a millimeter on a side. -
The optical guide sy~tem illu3trated in Fig. 19, or any `
~ariation thereof, can be utilized with any of the possible
EV generators and other components. Further, optical
reflectors such as the reflecting devices 156 and 158 can be
utilized with any other component. For example, a guide
system using tubular guides such as shown in Fig. 15 can
incorporate optical reflectors at the ends of the tubular
guide~.
j,
:i
3~0827
-35-
8. LC Guides
In general, as an EV approaches any circuit element,
the potential upon that element i~ depressed. The depressed
potential makes the element less attractive to the EV ~o
that, if there is a more attractive direction for the EV, a
steering action i9 available. Inductive elements are
particularly susceptible to the change in potential in the
presence of an EV, and thiq effect may be utilized in
providing an LC (inductance/capacitance) guide for EV's.
Fig. 20 shows an exploded view of a three-stage
quadrupole EV structure, indicated generally at 160 and
including three guide elements 162 mutually separated by two
spacers 164. Each of the guide elements 162 includes an
outer frame and four pole elements 162a, 162b, 162c and 162d
extending toward the center of the frame, but ending short
thereof to provide a central passage area. EV's, or EV
chains, enter the array of guide elements from one end of
the array, as indicated by arrow C, generally in a direction
normal to the plane of orientation of each of the guide
elements.
As illustrated, the four poles 162a-d are arranged in
mutually orthogonal pairs of opposing poles. There is
sufficient inductance in each of the poles to allow a
potential depression therein as the EV approaches. The
closer an EV passes to a given pole, the greater the
potential depression. Thus, for example, an EV approaching
closer to the lower pole 162a than to the upper pole 162c
causes a greater potential depression in the lower pole than
in the opposite, upper pole. The result is that the EV is
attracted more to the farther pole 162c than to the nearer
pole 162a. Consequently, a net force is applied to the EV
causing it to move upwardly, tending to balance the
potential depressions in the two opposed poles 162a and
162c. A similar result occurs in the opposed poles to the
sides, 162b and 162d, if the EV moves closer to one of these
poles than the other. Thus, a net restorative force urges
the EV toward the center of the di~tance between the two
opposed pole faces in eit;her the horizontal or vertical
~`
~33~827
-36-
directions. Any overshoot by the EV from the ¢enter portion
in either direction again unbalances the potential
depressions and causes a restorative force tendlng to center
the EV between the poles. It will be appreclated that the
net restorative force will also be generated if the EV
strays away from the center of the passage between the pole
faces in a direction other than horizontal or vertical,
causing unbalanced potential depre~sions among the four
poles so that such restorative force will always have
vertical and horizontal components determined by the
imbalance of potential between the opposed quadrupoles in
each of the two pairs.
Such restorative force tending to center the EV in its
pa~sage through a given guide element 162 may thus be
provided with each guide element. With an array of such
quadrupole guide elements 162, restorative forces will thus
be provided throughout the length of the array with the
result that the quadrupole element array acts as an EV
guide, tending to maintain the path of the EV centered
between opposed quadrupole faces. The spacers 164 merely
provide a mechanism for maintaining the quadrupoles of
adjacent guidance elements 162 separated from each other.
The entire array of guide elements 162 and spacers 164 may
be constructed as a laminar device, with guidance elements
in contact with adjacent spacers, for example. Further, it
will be appreciated that the LC guide of Fig. 20 may be
extended any length as applicable with additional guide
elements 162 and spacers 164.
An LC guide, such as that shown in Fig. 20, may be made
in a variety of shapes, and utilizing different numbers of
poles. In practice, the poles as illustrated in Fig. 20
resemble delay lines along the axis of a pair of opposed
poles. After an EV passes a set of poles, there will be a
rebound of the potential therein, depending upon the time
constant of the LC circuit. Eventually, the o~cillation~ in
the potential will subside. The timing function of the
guidance elements must be chosen to accommodate the passage
of subsequent EV's, for example. Further, it will be
, ~ " -.
' 3~7330827
appreciated that the LC guide of Fig. 20 operates without
the need of producing specific image-like forces, as in the
case of a dielectric of an RC guide, for correcting the
position of an EV as it passes therethrough, although the LC
guide mechanism can be con~trued as generating image forces
on a gross scale. Indeed, the guidance elements 162 and the
spacers 164 are conductors rather than dielectrics.
The coupling between the moving EV and the guidance
structure 160 dictates limits in the size of the structure
for a given EV size, that is, EV charge. If the guidance
structure 160 is too large in transverse cross section, for
example, the structure will not respond adequately to
control the EV; a too small structure will not allow
adequate turning time and space for the EV path to be
adjusted. Whether the guidance structure 160 is too small
or too large, its coupling with an EV will result in an
unstable mode of propagation for the EV and destruction of
the EV and damage to the guide structure. A factor that may
be utilized in the design of an LC guide 160 such as that
illustrated in Fig. 20 is to consider the poles to be
quarter wave structures at the approach frequency of the EV
to be guided. This frequency is determined primarily by the
velocity of the EV and the distance between the EV and the
steering, or pole, elements 162a-d. Since the diameter of
the guide 160 is related to the coupling coefficient, there
is an interrelationship between the diameter of the guide
and the spacing of the elements 162a-d. In this type of
guide, the quarter wave elements 162a-d can be operated at
dc or a fixed potential without charging effects. While an
LC guide can, in general, be made as large or small as
necessary to accommodate and couple to the particular size
EV's to be guided, the velocity range for propagation of
EV's to be guided by a given LC guide is not arbitrarily
wide.
It will be appreciated that the larger the number of
EV's in a chain to be guided, for example, the greater will
be the power level to be accommodated by the guiding
device. Generally7 an LV requiring an RC guide transverse
~`. .' ~ ' .
~; ~ . .
33~7
-38-
CKOSS section of 20 micrometerswould require an LC guide
- slightly larger. The spacing between the guidance
electrode~, or poles, such as 162a-d of Fig. 20, would also
be in the vicinity of 20 micrometers. Such sized elements
cannot be expected to handle very high power. Although
multiple, parallel units can be utilized to guide a flux of
EV's, it may be more economical of material use and
processing to scale up the EV structure to fit a larger
guide. Such scaling is primarily a function of the EV
generator or the charge combining circuits following the
generators when multiple generators are used.
The type of LC guide illu~trated in Fig. 20 may be
provided in many geometric and electric variations.
However, that type of structure is preferred for relatively
large sizes, and construction by lamination techniques.
Different construction techniques are applicable to smaller
structures and particularly to those amenable to film
processes. An exploded view of an LC guide made by ~ilm
construction is illustrated generally at 170 in Fig. 21.
The planar type LC EV guide 170 includes three guide
layers comprising an upper guide 172 and a lower guide 174,
and an intermediate guide system 176 interposed between the
upper and lower guides. The upper guide 172 comprises a
pair of elongate members 178 joined by cross members 180 in
a ladder-like construction. Similarly, the lower guide
includes longitudinally-extending members 182 joined
by cross members 184. The intermediate guide system 176
includes two elongate members 186 with each such member
having extending therefrom an array of stubs, or pole
pieces, 188.
With the three guide members 172-176 joined together in
laminar construction, the upper and lower cross members 180 ~:
and 184, respectively, cooperate with the intermediate
system pole pieces 188 to provide a tunnel-like passageway
through the array of cross members and pole pieces. In such
construction, the lateral confinement of the EV propagation
path is obtained by the conductive pole pieces 188
resembling quarter wavellength lines. The vertical
'-'' ::
133~2~
-39-
~ confinement, as illu~trated, is accomplished by the cross
¦ members 180 and 184, each operating as a shorted one-half
¦ ( wavelength line. The guide structure 170 effectively
operates as a form of slotted wave guide or delay structure.
Since the guide structure 170 is very active
electrically and can be expected to radiate strongly, the
structure may be enclo~ed with conductive planes on both top
and bottom to suppres~ radiation. Conductive radiation
shields 190 and 192 are illustrated to be positioned as the
top and bottom layers, respectively, of the laminar
construction. Since there is no fundamental need for
potential difference between the guide members 172-176, they
may be connected together at their edges, but, of course,
can be maintained isolated from each other with spacers if
desired.
In general, the EV's produced in a burst by most
generators are not highly regulated as to spacing between
the EV's, although in some instances, the spacing of
generated EV's can be affected. However, LC guides provide
some synchronization of EV's passing therethrough. The mean
velocity of EV's or EV chains passing through an LC guide is
locked to the frequency of the guide, and the spacing of the
individual EV's or EV chains is forced to fall into
synchronization with the structural period of the guide.
The resulting periodic electric field produced in the guide
tends to bunch the EV train within that field by
accelerating the slow EV's and retarding the fast EV's.
As the initial EV's move into an LC guide, there is a
short time period when the electromagnetic field level is
too low for strong synchronization. As the level builds up,
the synchronization becomes more effective. The "Q", or
~igure of m~rit of the guide as a cavity, determines the
rate of build up and decay. Too large a Q will cause
breakdown of the cavity. There is an implied optimum
filling factor for an LC guide as a synchronizer. With low
filling, the synchronization is not effective, and with high
filling, there is a danger of breakdown and interference
with the guide function.
. . :
~" ` ' :: - ' . ` ` ` ' ' . ' ' `
~ ~330827 ~
f ~
-40-
Better synchronization may be achieved when the
synchronizer is more loosely coupled to the EV's than the LC
guides of Figs. 20 and 21, for example. Such loose coupling
can be accomplished by using a slotted cavity providing
~mall slotq on one side of the guide. Then, the device
would operate at a lower frequency and have a much broader
pa~sband. Such a ~tructure is disclosed hereinafter a~ an
RF qource.
1330827 ~
-41-
; -
9. Surface Sources
Figs. 22-24 give three views of an EV generator
¦ comprising a surface source in con~unction with a guide
component. In general, guiding EV's on or near surfaces
requires coupling them from the source, or prior component,
to the surface in que~tion. In the case of a generator
I utilizing cathodes quch as illustrated in Figs. 4-6, for
example, it is possible to locate the source a short
distance from the propagating surface, and achieve workable
copuling. In the apparatus illustrated in Figs. 22-24, the
source of EV's is integral with the guide device along which
the EV's are to be propagated for enhanced coupling.
In particular, the generator and guide combination is
shown generally at 200, and includes a dielectric base 202
featuring a guide groove 204 and a surface, or planar,
cathode 206 embedded within the guide groove toward one end
thereof. A surface anode/counterelectrode 208 i5 positioned
on the opposite side of the dielectric base 202 from the
groove 204 and the cathode 206, and serve~ to effect
generation o~ the EV's and propagation thereof along the
groove. An optional top cover 210 is shown in Fig. 24 for
positioning against the grooved surface of the base 202, and
can be used without sealing provided the surfaces are
sufficiently flat. To avoid collecting charge in the
covered guide channel, the cover 210 is coated with a charge
dispersing material such as doped alumina, as discussed more
fully below.
In practice, the dielectric base 202 may be an aluminum
oxide ceramic plate or substrate with a thickness of
approximately 0.25 mm and a guide groove 204 with depth and
width approximately 0.1 mm each. The metallic coatings for
the cathode 206 and counterelectrode 208 may be of silver
paste compound fired onto the ceramic, for example. Mercury
may be wetted onto the silver cathode by applying the
mercury with a rubbing action. With such dimensions, the
operating voltage to produce EV's and propagate them along
the guide path 204 is approximately 500 volts. U~e of thin
film processing methods to produce a thinner dielectric
- . . .
. ~ .
~"',- . - ' .
~ . .
- ~4323~27
substrate 202 allows the operating voltage to be lower.
With such film technique3, aluminum oxide may be utilized
for the dielectric and evaporated molybdenum for the
metallic electrodes 206 and 208, all being depo~ited on a
substrate of aluminum oxide. In ~uch case, mercury can
still be used for migratory cathode material s~nce it can be
made to wet molybdenum by ion bombardment 3ufficiently for
such an application. Such bombardment may be by direct
bombardment of the molybdenum surface. Alternatively, argon
ion3 may be bombarded with mercury in the vicinity o~ the
molybdenum surface, thereby cleaning the molybdenum surface
for wetting. A small amount of nickel may be evaporated
onto the molybdenum surface to facilitate the cleaning of
the ~urface by direct or indirect mercury ion bombardment,
since mercury and molybdenum do not have high solubility.
The combination of molybdenum and mercury is preferred over
silver, or copper, and mercury because ~iluer and copper are
too soluble in mercury for use in a film circuit since they
can be rapidly di~solved away.
Since the cathode source 206 is effectively integral
with the dielectric substrate 202 in the guide groove 204,
the cathode is appropriately coupled thereto, that is,
transition of an EV from the cathode production region into
and along the guide groove takes place with minimal energy
loss by the EV. Additionally, the cathode 206, wetted by
mercury or the like, feature~ a self-sharpening or
regeneration action to maintain appropriately sharp its
leading edge, at which EV's are generated. Further, the
cathode 206 is an extended, or line, source so that pulse
repetition rates to produce EV's can be rai3ed to much
higher values than in the case of a single point source
becau3e the regeneration process involving migration of
liquid metal is not necessary between all pulses in the case
of an extended source as noted hereinabove. It will be
appreciated that the extended cathode 206 is identical to
the cathode 64, illustrated in Fig. 7, which is also mounted
directl~ on a ceramic base 62. Operation of such extended
cathodes relies on the frilnging field effects at the edges
~ 1330827 ~
43-
of the cathodes that cause a sharpening effect on the mobile
cathode wetting material. Consequently, one or more
relatively sharp structures can always be relied on for
field emission that i9 responsible for the EV initiation,
and there~ore the operating voltage of such a source is
relatively low.
330827 ~
1 ~ -44-
,
~1 lO. Surface Charge SuPpre~sion
~.
After an EV is generated, it may lose electrons due to
relatively poor binding of such electron~ at the time of
formation, or by some other proceqs such as pas~age of the
EV over a rough surface. In the latter case in particular,
the lost electron~ may di~tribute them~elve~ along the
sur~ace and produce a retarding field effect on subsequent
EV's pa~sing in the vicinity of the charged surface area.
Several technlques are available for removing this resulting
qurface charge.
The dielectric ~ubstrate, or base, employed in an EV
generator or RC guide, for example, experiencing the surface
charge buildup may be rendered sufficiently conductive so
that the surface charge iq conducted through the ~ubstrate
to the anode or counterelectrode. The resistivity of the
¦ base must be low enough to discharge the collected surface
charge before the passage of the next EV following the one
that charged the ~urface. However, the resistivity of the
surface cannot be arbitrarily low because the subsequent EV
would be destroyed by excessive conductivity to the anode or
counterelectrode.
To achieve the desired degree of bulk conductivity of
the substrate, the dielectric material, such a~ aluminum
oxide, can be coated with any of the resistant materials
commonly used for thick film resistor fabrication, provided
the resistance does not fall much below the range of 200
ohms per square. Such a resistive coating is usually
composed of a gla~s frit having a metallic component
included therein, and is applied to the surface by silk
~creening and sub~equent firing at an elevated
temperature. However, where intense EV activity occurs with
the utilization of high fields and possible high thermal
gradients, ~uch glassy material~ tend to break down and are
therefore unsatisfactory. In quch cases in particular, a
film of aluminum oxide doped with chromium, tungsten or
molybdenum, for example, may be added to the dielectric
component to provide a ~ufficiently conductive material,
thereby achieving the desired level of bulk conductivity of
. . ..
Z -- 133~27
-45-
the dielect~ic. The effectiveness of this procedure is
enhanced by decreasing the thickness of the substrate.
The photoemission spectrum from a decaying EV is rich
in ultraviolet light and soft X-rayQ if the di~turbance of
the EV causing the decay is severe. The absorption spectrum
of the produced photoconductor should be tailored to match
these high energy products. Since electron scatter and low
electron mobility in the photoconductor cauQes the
photoconductive proce~s to be slower than the passage of the
EV, the discharging of the surface charge due to the
decaying EV occurs slightly after the EV has passed a
particular location on the surface, and therefore poses no
threat of conducting the EV to the anode. In addition to
the ultraviolet and X-ray emission, part of the electron
emission from an EV near a surface excites fluorescence in
! the dielectric material, and the fluore~cent light then
contributes to activating the photoconductive process.
Another way of effecting surface charge suppression
through photoconductivity is by utilizing diamond-like
carbon for the dielectric component. Such material has an
energy band gap of approximately 3 ev, and thus can be
stimulated into photoconduction. Further, such carbon
material can be easily doped with carbon in graphitic form
to increase the conductivity of the substrate.
Another technique for dispersing the surface charge is
to utilize bombardment induced conductivity. Such
conductivity is activated by the high speed electrons coming
from the EV and penetrating a sufficiently thin layer
dielectric to bombard the anode, causing conductivity of the
dielectric applied to the anode. The -conductivity of the
- dielectric is effectively increased a3 the high velocity
electron stream is turned into a large number of low
velocity electrons in the dielectric. The dielectric
material is appropriately optimized for such proces~ by
1330827 ~
-45-
being sufficiently thin, with few trap sites. The trap
sites may be initially oleared thermally or optically, and
are cleared by the electric field during operation.
In general, the geometry of the dielectric substrate
may influence the effectivene~ of making the substrate
conductive to suppress surface charge, as in the cases of
photoconductivity and bombardment induced conductivity
techniques, for example.
~,, -
,,,
.;- ~,
.
~; 1433~827
11. Launchers
In some applications or structures, it is necessary or
desirable to propagate an EV acro~s a gap in vacuum or a
gaseous environment. For example, an EV may be launched
across a gap separating a cathode and an anode or guide
structure. The launching of an EV across a gap may be
accompli~hed by applying an appropriate voltage to attract
the EV from one region to the other. However, such an
applied voltage can represent a loss in power for the system
or the perhap~ unwanted energy gain for the EV. The
required applied voltage may be reduced to minimize the
system energy loss by inducing the EV to leave the cathode
region and enter into a counterelectrode region, for
example, without excessive energy gain. This may be
accomplished by propagating the EV across a region where the
field is high at the desired applied voltage so that the
field strips the EV from the surface along which it was
traveling and to which it was attached.
Fig. 25 illustrates a launcher construction, shown
- 20 generally at 216, designed to launch EV's across a gap
between an EV generator 218 and an EV guide, for example
220. The generator 218 includes a dielectric base which is
generally tubular, but closes at its forward end in a
conical structure ter~inating in a point 222. A
counterelectrode 224 is formed within the dielectric base by
conductor material coating the interior surface of the base
throughout the conical region thereof and extending partly
along the cylindrical portion of the base. A portion of the
exterior of the dielectric base is coated with conductor
material to form a cathode 226. The cathode 226 extends
along the cylindrical portion of the base and onto the
conical end of the base, but does not extend as far along
the base longitudinally as does the counterelectrode 224.
By terminating the cathode 226 short of the end of the
conical tip 222 the leading edge of the cathode, at which
EV's are formed, is maintained relatively close to the anode
j 224. Also, the truncated cathode 226 features a larger EV~
I producing area than woul~i be the case with the cathode
~330~27
-48-
extending to the tip 222 of the base. The fringing field
effect around the leading edge of the cathode 226 close to
the anode 224 is used in the production of the EV's. The
counterelectrode extends farther to the left within the
cylindrical portion of the base than the cathode coats the
cylindrical exterior of the base.
The tubular guide member 220, which is generally
constructed like the tubular guide illustrated in Fig. 15,
is coated on its exterior surface with conductor material to
form a counterelectrode 228 which extends throughout most,
but not all, of the length of the guide member. The
counterelectrode 228 does not extend to the ends of the
guide member 220 lest the EV's propagate onto the
counterelectrode. The end of the guide member 220 facing
the generator 218 features an internal conical surface 230
so that the generator tip 222 may be positioned within the
conical end of the guide member while still maintaining a
spacing between the two bodies. The guide member 220 may
also be constructed to circumscribe the generator 218,
provided the counterelectrode 228 is kept back from the
region of the cathode 226.
In operation, an appropriate potential difference is
applied between the cathode 226 and the counterelectrode 224
of tbe generator 218 to generate one or more EV's which
leave the forward end of the cathode and travel toward the
tip 222, under the influence of the field established by the
potential difference. It is intended that the EV's leave
the generator 218 and enter the interior of the guide member
220. Thereafter, the EV~s may propagate along the interior
of the guide member 220, under the influence, at least in
part, of the field established by the guide member
counterelectrode 228 generally as discussed hereinbefore.
The conical geometry of the generator end, and the relative
pos1tionlng of the generator cathode 226 and
.
,,
30827 (
-49-
counterelectrode 224 re~ult in the EV's experiencing a
large field at the generator tip 222 causing the EV's to
detach from the base of the generator 218. The EV's are
thus effectively eJected from the generator tip 222 at the
beginning of the guide member 220 and continue along, now
propagating under the influence of the guide member.
In practlce, the cathode 226 may be appropriately
wetted with a liquid metal conductor as discussed
hereinbefore. The guide member counterelectrode 228 may be
operated at the ~ame potential as the generator
counterelectrode 224, but other potentials can be u~ed. The
extraction voltage applied to the guide counterelectrode 228
is an inherent part of the generation proces~, and without
such voltage the generator will not produce EV's
effectively. The extraction voltage is normally ground
potential when the cathode 226 i5 run at some negative
voltage. With a negative-going pulse applied to the cathode
226 to generate the EV's, the generator counterelectrode 224
may be operated at ground potential. The mobile wetting
metal is drawn to a thin ring at the end of the cathode 226
nearest the tip 222. EV's are generated around the cathode
region so that, at a high pulse rate, there is a ~teady glow
around the cathode end accompanying EV production.
As an example of the construction of a launcher as
illustrated in Fig. 25, the dielectric body of the generator
218 may be made of aluminum oxide ceramic having a thickness
of 0.1 millimeter in the region o~ the conical end, that is,
at the wetted metal cathode edge, and being somewhat thicker
along the cylindrical shank of the base for additional
mechanical support. The counterelectrode 224 and the
cathode 226 may be fired on silver paste coating the
dielectric surface a~ discus~ed hereinbefore. Both the
¦ interior and the exterior of the conical end of the base 218
¦ are finely pointed to increase the field at the tip 222 to
cause detachmen-t of an EV a~ it approaches that region. The
spacing between the generator tip 222 and the nearest inside
surface of the guide member 220 may be on the order of
1 millimeter or le~s. With the foregoing dimensions, an EV
, , ::
~,,' . ;:, :,:' '., ' ,, ', ', . ' , : ~ ,
~$
133~827
;' may be formed and detached at the generator tip 222 with
approximately a 500 volt potential difference applied
between the generator counterelectrode 224 and cathode
226. A gas pressure on the order of 10-2 torr lifts the EV
off of the dielectric surface of the generator base 218 and
facilitate~ the transfer and propagation of the EV to the
guide structure 220, and even allows the cathode pulse to be
reduced to as low as 200 volts. High molecular weight
gases, such as xenon and mercury, are particularly good for
this function
It will be appreciated that the spacing between the
guide member 220 and the generator 218 may be ad~usted. In
a given application under vacuum or selected gaseous
conditions, requiring sealed operation, such movements can
be effected by a variety of techniques.
While a generally cylindrically symmetric launcher 218
is illustrated and described herein, it will be appreciated
that the launcher technique can be applied to EV generating
and manipulating components of any kind. For example, the
planar generator and guide illu~trated in Figs. 22-24 may
employ the launcher technique to overcome a large gap to a
subsequent guide member, for example, particularly when a
low voltage is utilized to generate the EV's.
In general, EV's may be formed and launched at lower
voltages if the dimensions of the components are
decreased. For low voltage operation, it is desirable to
use film coating methods to fabricate the components. For
example, to construct a planar launcher, an anode may be
formed by lithographic processes and then coated with films
of dielectric material such as aluminum oxide or diamond-
like carbon. After the deposition of the dielectric
material, the cathode material, typically molybdenum, can be
applied to the dielectric material, and then the entire
cathode may be wetted with a liquid metal. While a
generally cylindrical launcher may not be so fabricated
using film techniques, the electrodes may be painted on to
make such a launcher. ~ith dimensions of approximately
1 micrometer thickness fc)r the dielectric base of the
'
.
,,,
i
~ ~330827 ~`
~ 51-
generator, an EV may be formed and launched at a potential
difference between the cathode and anode of the generator of
le~ than 100 volt~.
Although the preferred embodiments of a launcher for
S EV'~ have been illu~trated and deqcribed herein, those
~killed in the art will realize that launcher~ for EV'~ may
be con~tructed in variou~ other form~.
.
~ 15
I
~r - ~ . ;. . .-; . ~ .
~330827 ~
;~ -52-
:
12. Selectors
A~ noted hereinbefore, EV's may be generated as beads
in a chain with multiple chains being produced at
essentially the ~ame time. It may be desirable, or
necessary, to iqolate EV's of a ~elected total charge for
use in a process or a device. A selector action can help
limit the number of types of EV's available to provide the
desired specie~. In general, a variety of EV's may be
generated and directed toward an anode or collector around a
sharp edge on a dielectric surface. An extractor field
detache~ selected EV's at the dielectric edge and propels
them toward a guide component or other selected region. The
extractor voltage a~ well as a guide voltage may be readily
¦ adJu~ted, in view of the geometry of the selector, to
extract EV's of a chosen charge size. Typically,
approximately five EV chains, each with ten or twelve beads,
may be extracted at a time, with the number of chains or
EV's scaled according to the geometry of the extracting
apparatus.
A generally cylindrically symmetric selector is shown
at 236 in Fig. 26, and includes a generator, or source, 238
constructed generally in the form of the separator shown in
Fig. 8. A generally tubular dielectric ceramic base 240 has
a conical forward end wherein the respective angles of taper
of the exterior and interior conical ~urfaces cooperate to
form a small aperture defined by a circular, ~harp edge
242. A conductive coating, such as a fired on silver paste
coating, forms a counterelectrode band 244 about the
exterior base of the conical end. A wetted metal cathode
246 is positioned within the tubular dielectric ba~e 240
with the cathode conical end within the conical structure of
the dielectric base and facing the aperture defined by the
edge 242. The cathode 246 may be copper wetted with
mercury, for example, as described hereinbefore.
An extractor 248, in the form of a conducting plate
with a circular aperture 250, i~ po~itioned in front of,
centered on and a short distance from the source circular
edge 242. Beyond the extractor 248 is a tubular guide 252,
~':
.
:
~ .:
1330827 1
-53-
for example, having a dielectric body with its external
surface coated, in part, with a conducting ~urface to form a
( counterelectrode 254.
If the generator 238 is operated to produce EV'~
without the application of a voltage on the extractor 248,
the EV's move from the region of the cathode tip to the
anode 244 by traveling through the hole in the end of the
ceramic cone and around the sharp edge 242 to the outside of
the cone and to the anode. When an appropriate voltage is
applied to the extractor, however, a selected portion of the
EV's at the dielectric edge 242 are detached from the
dielectric and propelled through the extractor opening 250
and to the guide member 252 through which they are
propagated under the influence of the potential placed on
lS the guide counterelectrode 254.
A planar selector is shown generally at 260 in Fig. 27
and include~ a generally flat dielectric base 262 having an
elongate neck 264. A surface source, or generator,
generally of the type shown in Fig. 22, is incorporated in
the selector 260 with a planar cathode 266 residing in a
groove 260. However, rather than being positioned on the
opposite side of the dielectric base 260, the anode used in
the generation of the EV's is in the for~ of a coating 270
on the side of a second groove 272 which inter~ects with the
first groove 268 at an acute angle to form a sharp
intersection edge 274. With a potential difference applied
only across the cathode 266 and the anode 270, EV's formed
at the cathode, which may be a wetted metal type, move along
the groove 268 to its inter~ection with the groove 272,
whereupon the EV's turn around the sharp edge 274 and
proceed to the anode 270.
Two extractor electrodes 276 and 278 are positioned
along the out~ide surfaces o~ the neck 264 of the base 262,
on opposite side~ thereof and flanking the guide groove
268. Application of an appropriate voltage to the extractor
electrodes 276 and 278 causes selected EV's negotiating the
sharp edge 274 to be detached therefrom and to proceed along
the guide groove 268 and through the region bounded by the
.,~, - : . ~ -
` ~ 133~827 ~
~ ~ -54-
"
~1 extractor electrodes. A~ shown in Fig. 28, a
~, counterelectrode 280 underlles a portion o~ the guide groove
268 along the neck 264 of the dielectric base to further
propel the selected EV's along the guide groove beyond the
extractor electrode3 276 and 278.
As noted hereinbefore, when an EV i9 traveling along a
surface, it is bound thereto by image force~. The magnitude
of the binding force depends to some extent upon the
geometry of the ~urface through which the image ~orce is
effected. When the effective area of the surface is
reduced, such a~ the case when an EV is passing about the
sharp circular edge 242 of the conical structure of the
generator 238 in Fig. 26, or about the qharp edge 274 of the
planar selector 260 in Fig. 27, then the image force is
reduced, and the EV becomes more loosely bound and sensitive
to being stripped away by a field provided by means of
another electrode with a relatively positive voltage applied
to it. The high negative charge of the EV's moving toward
the extractor electrode may momentarily reduce the potential
between the cathode and the extractor below the threshold
required to extract any of the remaining bead chains or
beads in the group at the edge in que~tion and moving toward
the source anode. After the initial EV structure i5
extracted and propagates beyond the extractor field, a
subsequent EV may be extracted from the region of the
dielectric edge.
As an example, in the configuration shown in Fig. 26,
¦ for an applied negative voltage of 2 kv on the cathode, an
aperture defined by the sharp edge 242 of approximately 50
micrometers, a cone radius of e~uivalent size, and a ~pacing
from the dielectric aperture to the extractor electrode of
approximately 1 millimeter, a positive extraction voltage of
approximately 2 kv is needed to detach an EV. The
extraction threshold voltage is critical. For example, when
an EV source of such dimensions is constantly firing and the
EV'3 are being captured entirely by the anode on the
dielectric cone, no extraction to the extractor occurs with
an extraction voltage of ]L.9 kv, but EV's are so extracted
F-
r~
i,.~ . ~ . .
., .
~L33~827 ~-~
-55-
at a poqitive extraction voltage of 2.0 kv.
While ~eparators are ~hown in Fig~. 24-26, a3
a330ciated with EV generators, 3eparator3 may be
incorporated virtually anywhere along a line of EV
S manipulating component~ For example, a ~eparator may
follow a guide device, or even:another separator. Providing
EV ~eparator~ in sequence, or even in ca3cade, permit3
extraction of EV'3 of a particular binding energy from EV's
i~ in a wide range o~ binding energies.
~ 15
"~
.'.
.
~.
":
;
.
. ,
i
~` 1 3 3 0 8 2 7
; -56-
1~. Splitters
y
In general, operations involving close timing or
synchronization of event~ can be controlled by two or more
output signals derived from a ~ingle input signal. For
5example, a first event can be divided into a multiplicity of
subevents. With an EV source that produces a large number
of EV bead3 or bead chains within a very short period of
time, $t i~ possible to divide such an event, that is, to
divide a burst of EV'q, into two or more EV propagation
10signals. Apparatuq for so dividing EV signals is called a
splitter, and is constructed generally by interrupting a
guide component, such as the RC guide devices illustrated in
Figs. 11-16, with one or more side guide channels
intersecting the main guide channel. As EV's mo~e along the
15main guide channel and reach the intersection of the main
channel with a side, or secondary, channel, some of the EV's
move into the secondary channel while the remainder continue
along the main channel. In constructing a splitter, care
must be taken to ensure that the secondary guide channel
20intersects the main channel at a position where the EV's
actually propagate. For example, if the main channel is
relatively large so that EV's may move along at a variety of
locations throughout the transverse cross qection of the
main channel, then there can be no certainty that an EV will
25encounter the intersection of the secondary channel with the
main channel sufficiently close to the secondary channel
entrance to move into the secondary channel.
A splitter qhown generally at 290 in Figs. 29 and 30
includes a dielectric base 292 with a mosaic tile 294 bonded
30to the base. A second tile piece 296 is also bonded to the
base 292. The tileq 294 and 296 are cut as illustrated and
bonded to the base 292 appropriately separated to form a
qecondary guide channel 298 between the two tiles. A ~ingle
tile, generally rectangular as viewed from the top in Fig.
3529, may be cut into two pieces to form the channel 298 when
the pieces are appropriately bonded to the base 292.
As discussed hereinbefore, a 90 angle between the edge
of quch a mosaic tile and the base 292 would form a channel
;~
.,~
''~
:
.~ , .
.`i ~ 1330827
-57-
to which EV's would be attracted and along which they would
be guided. However, providing a 45 bevel forms an acute
angle primary channel 300 when the tiles 294 and 296 are
~! bonded to the base 292, in the same fashion that such a
channel is provided by the guide member 110 illustrated in
Figs. 13 and 14. A guide counterelectrode or ground plane
302 for contributing to the attractiv~ force maintaining the
EV's within the guide channels is positioned on the opposite
side of the base 292 from the tiles 294 and 296. The
dielectric tiles 294, 296 and base 292 may be constructed of
any suitable material, ~uch as aluminum oxide. Similarly,
the counterelectrode 302 may be formed by any suitable
i conductor material, such as silver paste. The potential
applied to the counterelectrode 302 is chosen according to
`! 15 the application and other potential levels used, and may be
positive or ground.
A second version of a splitter is shown generally at
~j 310 in Fig. 31, and includes a dielectric base 312 with a
primary, straight guide channel 314 and a secondary guide
channel 316 branching off of the primary channel at an acute
angle. The channels 314 and 316 are grooves of rectangular
cross section formed in the base 312. As shown in Fig. 32,
a counterelectrode 318 is positioned on the opposite side of
the base 312 from the channels 314 and 316 to promote
propagation of the EV's along the channels, and a flat,
dielectric cover 320 is provided for optional placement
against the top surface of the base to enclose the guide
channels. In order to ensure that EV's moving from left to
right along the main channel 314, as viewed in Fig. 31, are
sufficiently close to the side of the main channel broken by
the secondary channel 316, it is necessary that the primary
channel cross section not be much larger than the mean size
of the EV's that are propagated along that channel, although
each channel has to be large enough to accommodate the
largest EV structure to be propagated therethrough. (The
mosaic guide channel with the bevel 300 in Figs. 29 and 30
will accommodate any size EV structure because it has an
open side.) Typically, for an EV bead chain formed at 2 kv,
~'-',. - , i, ,
133~827 ~
~ -58-
.,
the primary channel lateral dimension should be 20
micrometers. The lower limit for a channel width guiding a
single EV bead is approximately 1 micrometer. But, where EV
bead chains ~ormed at 2 kv are to be propagated along both
channels of the splitter 310, the width of the secondary
channel 316 qhould be at least 20 micrometers and the width
of the primary channel 314 may range between 20 micrometers
and 30-35 micrometerq.
Both splitters 290 and 310 may be utilized with a
variety of other components, and, for example, EV's may be
launched or propagated into the primary guide channels 300
and 314 from any of the sources disclosed herein. In the
case of the splitter 290 of Figs. 29 and 30, EV's or EV bead
chains move along the apex of the channel formation bevel
300 until the intersection with the secondary channel 298 is
reached. At that point, some of the EV's or EV bead chains
move into the secondary guide channel 298 and the remainder
continue to the right, as viewed in Fig. 29, along the
primary channel 300. The secondary channel 298 guides the
EV's or EV bead chains having entered that channel around
the elbow of that channel as illustrated, so that two
streams of EV's or EV bead chains arrive at the right end of
the splitter 290 as viewed in Fig. 29 along the two channels
300 and 298. From there, the EV's may be manipulated or
exploited by other components.
Similarly, EV's or EV bead chains launched into the
left end of the primary channel 314 of the splitter 310 of
Figs. 31 and 32 move along that channel until some of the
EV's or EV bead chains enter the secondary channel 316 and
are guided around its elbow so that two streams of EV's or
EV bead chains arrive at the right end of the splitter for
further manipulation or exploitation.
A single EV moving along the primary channel of either
of the splitters 290 and 310 illustrated may be expected to
turn into the narrower secondary channel in each case.
However, it is noted that a stream of EV's or EV bead chains
will be split as described, with qome of the propagation
following the main guide channel and the remainder following
~,~" ,~' ,, .
i`
5` ''''~ 33~827
; -59-
,,
~1 the secondary channel. The deflection of only a portion of
an EV propagation qtream into a secondary channel of a cross
( section smaller than or equal to that of the primary channel
may be due to a crowding effect of multiple EV's or EV bead
chains at the channel intersection, perhaps cau~ed by the
high concentration of charge of the EV's, that prevent~ the
total EV group from taking the secondary path. This i~ a
form of self-switching in which one or a few EV structures
pass into the secondary channel at a time while others
continue along the main path. In any event, splitters of
the type illustrated in Figs. 29-32 are effective in
producing multiple streams of EV propagation generated as a
¦ single stream from a single source. Additionally, the
arriva~ of the EV's at the output ends of the primary and
secondary channels are effectively simultaneous, since the
difference in path length along the primary and secondary
channels is insigni~icant. Consequently, multiple EV's
generated with a single signal pulse and arriving at the
junction of primary and secondary guide channels, for
~ 20 example, may split up with some EV's propagating along each
3 guide channel to produce EV arrivals, or signals, at two
locations. If the guide channel path lengths are identical,
the EV's may arrive at the end points of the channels
simultaneously, or nearly so.
A variable time delay splitter i9 shown generally at
330 in Figs. 33 ~nd 34 for use in producing a pair of EV
propagation signals, generated from a single burst of EV's
but arriving at a pair of locations at specified times which
may be essentially the same or different. The time delay
splitter 330 includes a dielectric base 332 to which are
bonded three mosaic dielectric tiles 334, 336 and 338. A
pointed cathode 340, such as those illustrated in Figs. l
and 2 or 17, is shown for use in generating EV's for
propagation along a first path 342 extending along the
intersections of the base 332 with the top edges (as viewed
in Fig. 33) of the two tiles 334 and 336. The path 342
further extends upwardly, as shown in Fig. 33, along the
intersection of the base 332 with the left edge of the
,...
. ~
~` :
~ 3 3 ~ 8 ~ 7
-60-
rectangular tile 338, along its upper edge and downwardly
along its right edge.
The first tile 334 is in the form of a trapezoid which
cooperates with the second tile, 336, which is in the form
of a triangle, to provide a channel 344 separating these two
tile~ and intersecting the primary path 342 at an acute
angle to form the initial leg of a secondary guide path 346.
A generally U-shaped dielectric tile 348, having left
and right legs 350 and 352 for extending about the lower
portion of the rectangular tile 338 as illustrated, is
movable, and may be selectively positioned, relative to the
rectangular tile 338 as indicated by the double-headed arrow
. The secondary path 346 continues downwardly, as viewed
in Fig~ 33, along the 90 intersection (see Fig. 34) o~ the
base 332 with the left side of the tile 338, until the path
reaches the tile leg 350. The movable left leg 350 has a
45 beveled lower inner edge 354, as shown in Fig. 34.
Consequently7 the secondary path 346, which follows along
the intersection of the base 332 and the left edge of the
rectangular tile 338 below the channel 344, is guided then
by the intersection of the base 332 and the beveled edge 354
of the leg 350 as the EV's prefer the more confined
intersection than the 90 interse-ction of the edge of the
tile 338 with the base 332. Consequently, the EV path 346
leaves the tile 338 to follow the tile leg 350. It will be
appreciated that the movable tile 348 may be positioned with
the leg 350 at the outlet of the channel 344 so that the
secondary path 346 follows the leg without first following
the left side of the tile 338. The secondary path 346
advances to the base of the U-shaped tile 348 and thereafter
moves across the tile base to the right leg 352, which
intersects along its left edge with the base 332 at a 90
angle as illustrated in Fig. 34. However, the lower right
~ edge of the tile 338 features a 45 bevel 356 as an
¦ 35 intersection with the base 332. Consequently, EV's moving
up~ardly, as shown in Fig. 33, along the intersection of the
tile leg 352 with the base 332, then move along the beveled
intersectlon of the tile 338 with the base, and upwardly
~ .
, : ~
:~ ,
,:
-" 1330827--
-61-
away from the end of the movable leg. As shown in Fig. 34,
a counterelectrode 358 underlies the base 332 to provide the
necessary potential for enhanc~ng the guiding effects of the
path~ 342 and 346 and, where the splitter 330 includeg a
cathode 340 for the generation of EV's, to provide the
potential for such generation.
I The right edge of the rectangular tile 338, as viewed
j in Fig. 33, includes two launchers 360 and 362 in the form
of dielectric extensions ending in sharp edges. Thus, EV's
moving along the 90 inter~ection of the upper portion of
the right edge of the tile 338 with the base 332 are guided
by the intersection of the launcher 360 with the base.
However, the launcher 360 is generally triangular in cross
section, as shown in Fig. 33, to provide a sharp edge at the
right end of the launcher. The EV will go forward onto the
flat substrate of the base 332 rather than turning around
the sharp corner of the launcher 360. This forward movement
of the EV is greatly influenced by the exact shape of the
leading edge of the launcher 360, which must therefore be
relatively sharp and straight to avoid launching EV's at
undesired angles. An external field may be provided by
electrodes (not shown) placed to the right of the launcher
360 for further manipulation of the EV's.
Similarly, the launcher 362 features a sharp edge
toward its right end so that EV's moving along the beveled
intersection of the lower right edge of the tile 338 with
the base 332 turn toward the right, as viewed in Fig. 33, to
move along the perpendicular inter~ection between the
launcher 362 and the base, and then out over the base away
from the launcher. EV's exiting the launcher 362 may be
further manipulated by an appropriate external field applied
with the use of appropriate electrodes (not shown).
The primary path 342 is a fixed path, that is, it has a
singular path length between the intersection of that path
and the channel 344, for example, and the launcher 360. On
the other hand, the secondary path 346 is variable in path
length between the intersection of the channel 344 with the
primary path 342 and the second launcher 362, for example.
- ' 133~g2~ '
--62-
This variation in path length is achieved by movement of the
U-shaped dielectric member 348 relative to the rectangular
( tile 338 as indicated by the double-headed arrow E. The
farther the dielectric member 348 is positioned downwardly
S relative to the tile 338, as viewed in Fig. 33, the longer
will be the secondary path length 346 ( and the shorter will
be the overlapped portions of the legs 350 and 352 with the
respective sides of the tile 338). By selectively
positioning the dielectric guide member 348 relative to the
tile 338, the length of the path 346 may be selected and, in
this way, the time required for EV's to traver~e the
secondary path 346 and arrive at the second launcher 362 may
¦ be chosen. Consequently, the relative time of arrival at
the two launchers 360 and 362 of EV's generated by a single
pulse, for example, and following the two paths 342 and 346
may be selected by the positioning of the dielectric guide
member 348.
The 10 mm dimension indicated in Fig. 33 shows a
typical scale for a variable splitter. It will be
appreciated that differences in path lengths on the order of
a tenth of a millimeter or less may be readily effected
using a variable splitter of the size indicated. Any
appropriate means may be utilized to move and selectively
position the movable guide member 348, including a
mechanical linkage for example. If necessary, where the
adjustment is made manually, a form of micromanipulator or
translator, such as a lever and/or gear system with
appropriate mechanical advantage may be utilized to achieve
the desired sensitivity of control.
~t will be appreciated that the guide paths 342 and 346
may be modified as appropriate to any application. Further,
the paths need not extend to launchers 360 and 362, but may
continue on to further guide paths, for example, or other
components as appropriate.
For example, a version of a variable time delay
splitter ~ shown generally at 370 in Fig. 35. The
construction and operation of the splitter 370 is similar to
that of the splitter 330, and need not be further described
ii
l 133~827(
~63-
in detail, except for the differences therebetween. For
example, the fixed guide path 372 may be the same a~ the
fixed guide path 342 in Fig. 33, but the variable guide path
374 provided by the splitter 370 i3 ad~usted by a movable
guide member 376 (a~ indicated by the double-headed arrow F)
which extendQ farther to the right, as viewed in Fig. 35,
and ends in a launcher 378 whi~h expel~ the EV'~ along a
line directed toward a point of intersection, G, with the
first guide path 372. Thu3, EV'~ may be cau~ed to reach the
point G from two different direction~ at the same time, or
at selected different times, depending on the po~ition of
the movable guide member 376. WitneQs plates, or other EV-
¦ detecting devices such as phosphorous screens, 380 and 382
may be positioned to receive the EV's moving along the
primary and secondary paths 372 and 374, respectively.Additionally, appropriate anodes or counterelectrodes may be
utilized to enhance or further the movement of the EV's from
the launchers.
In general, the secondary channel of a splitter may be
larger, smaller or equal in transverse dimensions to the
main channel. If the secondary channel is much larger in
cross section than the primary channel, all EV propagation
may follow the secondary channel. The secondary channel
may intersect the main channel at any acute angle up to
90. The channels may mutually branch in various
patterns, such as to form a Nyl~ or a "T", for example.
For such examples, the two branches may be equivalent
channels. Further, multiple secondary paths may be
utilized so that any number of output signals may be
constructed from a single input EV signal from a single
source, for example. It will be appreciated that
splitters may also be constructed in forms different from
those illustrated in Figs. 29-35. For example, splitters
may be constructed utilizing generally tubular guide
components as discussed hereinbefore.
. . ~ . - .
l~ ~
~330827 ~
-64-
14. Deflection Switches
As noted, not only may EV's and EV chains be propagated
in selected directions by use of guide components, but the
guide components may also include turns in the guide paths
to selectively change the direction of propagation. The
guide components influence the direction of propagation of
EV'~ due to the attraction EV's experience toward the
dielectric guide ~urfaces caused by image charge forces on
the EV's, as well as the fields established by counter-
electrode9 further attracting the EV's to the dielectric
guide surfaces. The direction of propagation of EV's and EV
bead chains may also be influenced by the use of transverse
electric field~ acting on the electric charge of the EV
entities to deflect them to new, selected directions. The
extent of the deflection will depend on the size of the
deflecting field as well as the period of time over which
the field is applied to the EV entity. Additionally, the
deflecting field can be turned on or off, or set at varying
strengths to selectively deflect EV's differing amounts, or
not at all, as the EV's traverse a particular region. Of
course, there is a bilateral effect present, and the
deflecting mechanism, whatever form it may take, may
experience undesirable reaction from a countervoltage caused
by the EV passage.
As EV's move along guide paths, such as provided by
guide grooves as previously described for example, the EV
propagation path is very stable, not only due to the
potential well the EV's are traveling in due to the
dielectric image charge and counterelectrode field, but also
to the transverse wall boundaries established by the
dielectric groove in two or more transverse directions. In
order that an EV, moving along a guide channel, may be
deflected sideways by an applied field to a new direction of
propagation, the guide constraints in the direction of
deflection must be sufficiently low to permit the deflection
under the influence of a deflecting field. At the least,
the region in which deflection i~ to occur mu~t be free of
any guide channel wall that would interfere with the
~i~ 1330827
-65-
transverse deflection of the EV. In general, an EV moving
along a guide channel and experiencing a highly ~table
propagation path must be exposed to a relatively unstable
path in the region of the deflection; after the desired
deflection has occurred, the EV may again enter a relatively
highly stable propagation path along a guide channel, for
example. Where a choice is permitted, the EV may proceed in
one of two or more available post-deflection propagation
paths, depending on the application of a deflection field.
A device which is thus used to selectively change the
direction of propagation of an EV or EV chain, for example,
is a deflection switch.
Figs. 36-38 illustrate top, side and end views,
respectively, of a deflection switch shown generally at
~ 15 390. The EV deflection switch 390 is a single pole, double
a1; throw switch, constructed with a dielectric base 392
incorporating a single input guide channel 394 and first and
second output guide channels 396 and 398, respectively. The
input and output channels 394-398, which are shown as
mutually parallel but may be set at virtually any angles
relative to each other, are connected by a transition, or
deflection, region 400 which has the same depth as the guide
channels but which is generally broadened. A guide
counterelectrode 402 underlies the input channel 394, and
guide counterelectrodes 404 and 406 underlie the output
channels 396 and 398, respectively, for the application of
appropriate voltages to enhance the propagation of EV's
along the respective guide paths.
Two deflection electrodes 408 and 410 are also
positioned on the bottom side of the base 392 oppos ite the
~ guide channels 394-398 and the transition region 400, the
¦ deflector electrodes extending laterally from positions
partly underlying the transition region outwardly to provide
relatively large surface area electrodes. Thus, an EV
entering the transition region 400 from the input guide
channel 394 may be deflected to the left (a~ viewed from the
point of view of the EV entering the transition region) by a
positive charge placed on the left deflector electrode 408
~33~8~7
-66-
and/or a negative charge placed on the right deflector
electrode 410. In this way, the path of propagation of
the EV is turned from the generally straight line path
~i enforced within the input guide channel 394. By
appropriate application of charge to the deflector
electrode 408 and/or the deflector electrode 410, the EV
path may be deflected so that the EV enters the first, or
left, output guide channel 396 along which the EV may
continue to propagate. Alternatively, charge may be
placed on one or both of the deflector plates 408 and 410
¦ to deflect the path of propagation of an EV emerging from
the input channel 394 so that the EV enters the second, or
right, output channel 398, along which the EV may continue
to propagate.
The deflection switch operates by allowing an EV to
move from a relatively highly stable path in the input guide
channel into a region of relative instability within which
the path may be selectively deflected by the application of
a deflector field, whereupon the EV may enter an output
guide channel providing another relatively highly stable
propagation path. The ~ransition from the input guide
channel to the transition region should be done in a manner
that does not set up transients in the EV path, otherwise
spurious switching can result. Feedback from the de~lected
EV can be used to completely relieve the effects of input
loading or coupling. For example, any nearby electrode will
pick up voltage feedback as an EV passes; the feedback
signal can be communicated to a deflection plate through an
appropriate variable amplitude, phase inverter coupling.
Those skilled in the art will recognize this as a push-pull
device. By reversing leads, it can be used to provide cross
coupling. Such a feedback electrode 412 is shown positioned
on the top of the base 392 ad~acent the left output channel
396 and connected by an appropriate lead to a coupling
circuit 413, the output of which is connected to the left
side deflection electrode 408. A similar feedback electrode
414 is positioned on top of the base adjacent to the right
output channel 398 and connected to a coupling circuit 415,
the output of which is connected to the right side
-
~33~27
-67-
deflection electrode 410. In this way, degenerative or
regenerative feedback may be achieved to produce a stable or
unstable, that is, bistable, switching process,
respectively Other known feedback effects may be achieved,
with a different feedback circuit for each effect.
Similarly, filters can be constructed with the feedback
circuitry to limit the switching of EV's to an output
channel according to charge magnitude or other parameters,
for example. There iq a considerable advantage in having
the feedback circuit use electromagnetic components
operating near the velocity of light to circumvent the
delays that would otherwise produce poor transient
response. Conventional resistor, capacitor and inductance
components in general work well with EV's traveling at about
O.l the velocity of light.
The deflection switch 390 illustrated in Figs. 36-38
may be constructed by etching the guide paths and transition
region into fused silica using photolithographic techniques,
for example. The conductive electrode deposits can be made
using vacuum evaporation or sputtering methods. The depth
and width of the input and output guide channels should be
approximately 0.05 mm for operation with EV's generated at
about 1 kv. The deflection voltages applied to the
deflector electrodes may range from tens of volts to
kilovolts, depending upon the degree of stability of the
path of the EV passing through the transition, or
deflection, region. The degree of stability of the EV path
within the transition region depends upon the shape and
length of the transition region as well as the
configurations of the counterelectrodes.
To optimize the deflection sensitivity of a switch, the
EV propagation path should be more unstable down the middle
of the tran~ition region. For example, the deflection
switch 390 features a transition guide portion 400 with side
walls 416 which intersect the input channel guide walls at
right angles to mark an abrupt end of the input guide
channel 394. Such an abrupt mechanical transition requires
high deflection voltages to selectirely control and deflect
: :-
. ~ ,
!;1,
~ ~330827 ~
-68-
the EV's within the transition region since the EV's can
merely lock onto one of the side walls of the tran~ition
guide region 400, opposite to the desired deflection
direction. Consequently, high deflection voltage would be
required to switch an EV across the transition guide section
400 to the opposite wall.
The transition from the input channel 394 to the
deflection guide area 400 can be made more gradual, and the
deflection sensitivity of the device increased, by
particularly patterning the electrodes, including the input
guide counterelectrode 402. For example, as illustrated,
the input guide counterelectrode 402 does not end at the
intersection of the input guide channel 394 with the
intermediate transition section 400, but rather continues on
in a tapered portion 418 extending partly under the
intermediate section. Accordingly, the deflector electrode~
408 and 410 are truncated to parallel the tapered portion
418 of the input counterelectrode 402. Such an electrical
transition technique allows an EV to move from the input
guide channel 394 to the intermediate guide section 400 with
little disturbance, that i~, with no significant change in
propagation path in the absence of a deflector field,
thereby promoting high deflection sen~itivity. Without the
use of a counterelectrode in general, the EV propagation
path cannot be readily predicted.
As illustrated, the intermediate region 400 forms a
~hallow V-shaped wall 420 between the fir~t and second
~ output guide channels 396 and 398, respectively. The shape
¦ of this portion 420 of the intermediate guide section side
~ 30 wall is relatively ineffective in controlling the stability
! of the EV paths within the intermediate region.
¦ Alternatively, an EV may be introduced into the
intermediate transition section for deflection with low
disturbance with the use of a mechanical design to provide a
gradual tran~ition of the EV from the influence of the input
guide channel to the intermediate guide region. For
example, such a deflection switch may feature an input guide
groove which tapers in the thickness direction, or depth, in
,~ ,r_
-~ 1330~27
-69-
i
con~unction with an input guide counterelectrode which ~ay
end relatively abruptly, and may even be ~quared off, for
example. For example, a tapered top surface 422 about the
input channel 394 is shown in phantom in Fig. 37 as an
S illustration of such mechanical design. The input guide
channel gradually loses its effectiveness in guiding the EV
as the EV advances toward the deflection region, thus
negotiating a transition between the two regions with little
disturbance of the propagation path of the EV in the absence
of a deflector field, and again providing relatively high
deflection sensitivity. It will be appreciated that etching
techniques in general yield tapered edges rather than
abrupt, squared-off edges at the ends of surfaces. This
naturally occuring etch taper may be exaggerated to achieve
the taper such as illustrated at 422 in Fig. 37.
A technique to give greater stability against charge
collection is to use a low resistance coating for the
deflector electrodes, and placing these electrodes on the
upper surface within the transition region 400 rather than
under the region. Thus, the EV path will generally cross a
deflector electrode. Dielectric charging is prevented by
using this deflection method.
.,
~-`.: , - ~ ~ , . -
~s . . .
:~'
~330~27
-70-
15. EV Oscilloscope
t
An EV or EV bead chain traveling acro3s a surface in
( vacuum may do 90 in an erratic fashion due to local fields
and surface disturbance3. Such movement i~ accompanied by
the e~ection of electron~ from the E~ so that it~ path is
vi~ible when viewed by an electron imaging sy~tem or by the
e~ected electrons ~triking a nearby phosphor that produces
visible light. By utilizing field forming structures, such
as deflection electrodes, to impre~s electric field~ to
control the path of an EV, the path, and therefore its
optical image~ can be made to describe the time varying
function of the applied voltage, thus providing the
functions of an 03cillo3cope. This can be effectively
achieved by extending the quality of the stabilizing and
deflection methods of the EV switch 390 of Figs. 36-38.
An EV oscilloscope of the planar type is illustrated
generally at 424 in Fig. 39, and includes a dielectric
` substrate, or base, 426 featuring an EV input guide channel
` 428 opening onto a flat transition, or deflection, area 430
after the fashion of the tran~ition area 400 of the
deflection switch 390 in Fig. 36. A guide counterelectrode
432 underlies the guide groove 428, but ends in an extended
¦ taper under the deflection area 430 as illustrated. The
leading wall 434 of the deflection area 430 is set at a 90
angle relative to the input channel 428. Consequently, the
combination of the tapered counterelectrode 432 and the
structure of the deflection area wall 434 relative to the
input channel 428 maximizes the stability of EV's or EV
chains entering the deflection area from the input channel
a~ discus~ed hereinbefore in connection with the deflection
~witch 390.
Two deflector electrodes 436 and 438 are provided on
the underside of the ~ubstrate 426 as illustrated to
selectively apply a signal to act on EV's moving acros~ a
selected portion, the active area indicated by the broken
line H, of the tran~ition area 430. The entire interior
area of the transition region 430 may be coated with
reslstive material to ~uppres~ ~urface charge and act as a
,.~
,
33as27
-71-
terminator for the transmission line feeding in the
;I deflection signal to the deflection electrodes 436 and
( 438. The bottom ~urface of the deflection area 430 must be
smooth to avoid local unintended structures which might
deflect an E~. The EV, or EV chain, propagate~ out of the
active area H and the deflection region 430 in general, and
may eventually be caught by a collector anode (not shown).
Fig. 40 i~ an end view of the EV o~cillo~cope 424,
showing the addition of a phosphor screen 440. The ~creen
440 is to be positioned over at least the active area H, but
~ may extend over the entire transition area 430 or even the
¦ entire substrate 426 aq illustrated. Electrons emitted from
the EV or EV chain moving under the influence of the applied
deflection field interact with the phosphor 440 to emit
light. An optical microscope 442 is positioned to receive
light emitted from the phosphor 440 for magnification and
observation. A light intensifying television camera can
al~o be used in this configuration in place of the optical
microscope. Magnification for the viewing system, whether a
microscope or a television camera, should be sufficient to
show an object of several micrometers, the approximate size
of an EV. Utilizing a television monitor to view the
activity o~ the oscillo~cope provides both increased
sen~itivity and easy recording ability. Additionally, an
electron camera, described hereinafter in Section 16, can be
utilized to look directly at an EV traveling on the
transition area 430, or even in space.
Any EV source compatible with launching into guides can
be utilized with the EV oscilloscope 424. If appropriate, a
separator or selector may also be utilized to provide the
desired E~ or EV chain entering the ~cope guide channel
428. Typically, the formation and launching voltage used to
obtain EV'~ for the oscilloscope 424 may range between 200
volts and 2 kv depending upon the size of the structures
utilized. As in the case of the deflector switch 390 of
Figs. 36-38, the design of the guide channel 428 (such as
its length) and counterelectrode 432, and the deflectlon
reglon 430 must be such a~ to provide a stabilized EV
D,., ~
~il
-~ 1330827 .~
-72-
launched into the deflection region 430 without locking onto
the side walls of the deflection region. The scope 424
effectively operates, in part, as an analog-type of switch
with many output states that are determined by the voltage
applied to the deflector electrodes 436 and 438.
The velocity of the EV moving out of the guide channel
428 and across the deflection region 430, coupled with the
image magnification provided by the optical microscope,
television system or electron camera, for example, represent
the horizontal scan rate of the oscilloscope 424 while the
electric field impressed orthogonally to this motion, by use
of the deflector electrodes 436 and 438, displays the
vertical axls. The EV motion resulting is not a true
function of the potential impres~ed upon the deflection
electrodes 436 and 438, but rather an integral of the
function.
Synchronization of the EV trace with the electrical
event being analyzed by use of the scope 424 may be
accomplished by generating the EV's slightly before the
event is to be displayed, as is usual for oscillography.
The sensitivity and sweep speed of the scope 424 may be
varied by changing the entire device geometrically, or at
least viewing a longer EV run in an extended active area H
for longer sweep times. Typically, the distance between
nearest points of the two deflector electrodes 436 and 438
may be in the range of approximately 1 millimeter, and
impressed ~ignal frequencies on the order of 100 GHz may be
utilized. The voltage range of the display is determined by
selecting a particular attenuation for the signal before it
is impressed upon the deflection electrodes 436 and 438.
Due to the small size of the EV and its relatively high
velocity, the bandwidth of an EV oscilloscope is relatively
large. Single event ~aveforms can be analyzed when the
transition times lie in the 0.1 picosecond range. Such a
fast o~cilloscope provides a significant tool in analyzing
high speed effects obtained with use of EV's. For such wide
bandwidths, as is possible with the "picoscope," it is
necessary to compensate th~e attenuators used in the signal
1330827 ''~5'
73
input circuitry to the deflection electrodes 436 and 438.
Use of microstructures in constructing the EV scope avoids
excessive signal time delayq. The scope 424 and any
associated circuitry should be operated as closely as
5 pOS9 ible to the electrical event being measured to prevent
dispersion in the coupling transmission line~. For much of
the work in the range of an EV scope, the scope may be
effectively embedded in the region generating the signal.
The picoscope essentially become~ a "chip scope," and may be
considered practically dispo~able.
I
~`.'' ' ~ '' ~ ' :
a~
~ 133~27 ~
-74-
! 16. E ectron Camera
As noted hereinbefore, an electron camera may be
utilized to view the electron emissions from EV's moving on
an EV oscilloscope, such as the picoscope 424 o~ Figs. 39
5 and 4O. Such an electron camera is qhown generally at 450
in Figs. 41 and 42. The camera 450 includes a metallic
casing 452 which serves as an electrical shield against
stray field~ which might otherwise affect the manipulation
of charge within the casing. A pinhole aperture 454 is
provided as an entrance to the casing 452 to allow
electrons, ions, neutral particles or photons, to enter the
casing while assisting in screening out stray charge, for
example. Typical scale for the camera 452 is indicated by
the 25 millimeter dimension shown in Fig. 42. Typical
lateral dimension of the aperture 454 is approximately 50
micrometers.
~; A pair of deflector plates 456 and 458 are positioned
within the casing 452 so that charged particles entering the
aperture 454 are generally directed between the deflection
plates. Terminals 460 and 464 extend from the deflection
plateR 456 and 458, respectively, through the wall of the
casing 452 and are insulated therefrom by insulation shafts
462 and 466, respectively. A combination channel electron
multiplier (CEM) and phosphor screen 468 is pos itioned
across the end of the casing 452 opposite the aperture
454. Charged particle~ impact the CEM, which produce~ a
cascade effect to yield a magnified charge impact on the
screen, which glows to optically signal the original impact
on the CEM at the location opposite the glow on the
screen. The construction and operation of such a CEM and
phosphor screen combination 468 are known, and need not be
further described in detail herein.
The casing 452 is open at the phosphor screen, except
with the possible addition of a conducting ~ilm to complete
the ~hielding provided by the casing, but which will not
interfere with the emerKence of light from the pho~phor
screen to be viewed outside the casing. Although not shown
~33~27 1
-75-
in the drawing~, the CEM and pho~phor ~creen 468 are
provided with appropriate lead connections by which ~elected
( voltage3 may be applied thereto separate from the potential
at which the casing 452 may be set, and by which a potential
difference may be effected between the CEM and the phosphor
screen. Typically, the potential difference between the CEM
and the pho~phor screen is 5 kv, while the CEM gain i~
independently varied by setting it~ potential. In general,
the variou~ componentq of the camera 450, including the caRe
a 452, may be ~et at either polarity and at any potential, at
lea~t up to 5 kv.
In addition to the capability of having various
voltage~ applled to the casing 452, CEM and pho~phor screen
468 and electrodes 456 and 458, the camera 450 may also be
mounted for ~elected movement and positioning relative to
whatever is being examined by mean~ of the camera. Thus,
; for example, it may be appropriate to move the camera
- longitudinally and/or sideways~ or rotate the camera about
any of it~ axes.
Charged particles, such as electrons, entering the
aperture 454 may strike the CEM 468 at any point thereof,
with the result that a bright spot is produced on the
phosphor screen and can be viewed as an indication of some
event. The deflection plates 456 and 458 are provided for
use in performing charge or energy analysis, for example,
or in other measurements. Retarding potential methods,
utilizing the voltage on the CEM, for example, may also be
used in the analyses. Such analysis techniques are known,
and need not be described in detail herein.
The pinhole camera 450 has a variety of application~ in
conjunction wi~th EV's, for example. In Fig. 41, an EV
source 470 and anode 472 are pos~tioned in front of the
camera aperture 454 80 that EV's may be extracted from the
source and passed through an aperture in the extracting
anode. The EV's will ~trike th-e front of the camera 450
around the aperture 454, which may be in a molybdenum
plate. A brass ring tnot shown) may be placed in front of
the plate with the aperture 454 to receive the EV'~ and
.1''
~l
`` ` ~330827
-76-
,
prevent them from striking the face of the camera 450. A
metal foil may be placed across the aperture 454 to serve as
a target. In another ~uch arrangement, the combination of
the EV ~ource 470 and the extractor 472 may be positioned at
a different angular orientation relative to the camera 450,
such as at 90 relaSive to the configuration illustrated in
~ Fig. 41 so that generated EV's are made to pass by the
i camera aperture 454 with the re~ult that some electrons
emitted from the passing EV may enter the camera aperture
for observation of the EV propagation.
Fig. 43 illustrates how the camera 450 may be used in
~ con~unction with an EV oscilloscope such as the picoscope
¦ 424 of Fig. 39. As illustrated in Fig. 43, the camera 450
may be positioned facing the active area H of the
lS oscilloscope 424 with the camera aperture a short distance
therefrom so that electron emission from an EV being used to
trace a signal on the scope active area may enter the camera
through the camera aperture and be detected by the CEM and
~ pho~phor screen. For such use of the camera, the deflection
i 20 plates 456 and 458 may be maintained at ground potential,
for example, while the CEM is maintained at sufficient
! voltage to accelerate the EV-emitted electrons to strike the
CEM. The lens system of a television camera 474 is
illustrated facing the light output end of the camera 450 in
Fig. 43. The CEM and phosphor screen combination already
provides a magnification of approximately 5 in the camera
450 as illustrated. The overall magnification of the
combination of the electron camera 450 and the television
camera 474 may be increased by use of the television
system.
Fig. 44 shows yet another use of an electron camera
450, here in conjunction with a second electron camera 450'
positioned so that the longitudinal axes of the two
cameras are mutually perpendicular and may be in the same
plane. In this way, the location of an EV, for example,
passing in front of the two cameras may be determined in
three dimensions. As illustrated, the cameras 450 and 450'
are positioned along the :K and y axe~, respectively, of an
3 3 0 ~ 2 7,
-77-
r
orthogonal xyz coordinate system, with the cameras "looking
back" toward the origin of the coordinate system. Two sets
of deflecting electrodes, including electrodes ll76 and 478
located in mutual oppo~ition along the x axis, and
electrodes 480 and 482 al~o located in mutual opposition and
along a line perpendicular to the axis of orientation of the
first pair of electrode~ 476 and 478, that is, along the y
axis, may be positioned as illustrated to ~electively
deflect an EV in the combined field of view of the cameras
450 and 450 ' . The electrodes 476-48~ may be thin wires, say
on the order of 0.5 mm in diameter, so that the wires 478
and 481 neare~t the cameras 450 and 450', respectively, may
be placed in front of the respective cameras without
interfering with the line of sight of the cameras, that is,
the cameras "see sround" the wire electrodes. Appropriate
¦ leads to the electrodes 476-482 permit setting them at
desired potentials. In this way, as noted hereinbefore in
the discussion of an EV oscilloscope in Section 15, an EV
oscilloscope operating in three dimensions can be
constructed and utilized with two electron cameras.
Fig. 44 also illustrates the use of a third electron
camera 450" positioned along the z axis, for example, to
further observe the behavior of EV's in three dimensions in
conjunction with the x and y cameras, 450 and 450~,
respectively. Field electrodes 484 and 486 are provided
along the z axis to deflect EV's in that direction.
Two electron cameras may be positioned along the same
line, such as cameras 450" and 450 shown in Fig. 44
facing each other along the z axis, to perform Doppler
energy analyses on electrons, for example.
As in the case of the picoscope of Section 15, for
example, any appropriate EV source, with EV manipulating
components disclosed herein, may be utilized to introduce
EV's into the field of observation of any of the camera
arrangements indicated in Fig. 44.
....
;~' ' ` .
.
~ ~30827
~ 78
b
~; 17. Multielectrode Sources
The ~eparators, selectors and launchers described
hereinbefore are forms of multielectrode sources, or EV
generators, designed for specific purposes as noted; that
is, these device~ include electrodes in addition to a
cathode and single anode, or counterelectrode used to
generate EV's. Multielectrode devices may be used for other
purposes as well. For some applications, it may be
necessary to maintain a fixed cathode and anode potential
difference for EV generation while still exerci~ing
selective control over the production of EV's. This may be
accomplished by adding a control electrode to form a
triode. One version of a triode source is shown generally
at 490 in Fig. 45. The triode 490 is constructed on a
dielectric base 492 featuring an elongate guide groove 494
in which is located a planar cathode 496. An anode, or
counterelectrode, 498 i5 positioned on the opposite side of
the base 492 from the cathode 496, and toward the opposite
end of the base. A control electrode 500 is also positioned
on the opposite side of the ba~e 492 from the cathode 496,
but closer longitudinally to the end of the cathode then is
the anode 498. Effectively, the control electrode 500 is
! pos itioned between the cathode 496 and the anode 498 so that
the voltage of the control electrode may significantly
affect the electric field at the emission end of the cathode
¦ where EV's are formed.
With fixed potentials applied to the cathode 496 and
anode 498, an EV may be generated at the cathode by pulsing
the control electrode 500 in a positive sense. There is a
sharp threshold for effecting field emission at the cathode,
the process that initiates the generation of an EV.
Therefore, a bias voltage may be applied to the control
electrode 500 with a pulse signal of modest voltage
amplitude to generate EV's. In such case, no dc current is
drawn from the control electrode 500, but large ac currents
are present with the pulsed signal.
A triode operates by raising the cathode emission
density to the critical point required for the generation of
~ 1 33 ~827 ~
~ 79-
;
an EV. As in triodes in general, some interaction between
the control electrode 500 and the output of the source 490
may occur. The control electrode 500 must be driven hard
enough to force the first EV and a subsequent EV into
5 existence because of the strong feedback effects that tend
to quppre~s the creation of the EV~s. Standard feedback at
high frequencies diminishe~ the gain of the generatGr, so
that the control electrode cannot be raised to sufficiently
high positive potential to effect ~ubsequent EV
10 generation. For example, a~ the control electrode voltage
is being raised in a positive ~ense to effect initial EV
generation at the cathode 496, the capacitance of the
combination of the control electrode and the anode 498
increases due to the presence of an EV as well as the
15 increase in the control electrode voltage. When the first
EV formation begins, the effect of the control voltage is
reduced due to space charge. As the EV leaves the region
over the control electrode 500 and approaches the region
over the anode 498, there i~ a voltage coupled to the
20 control electrode that depend3 upon the anode instantaneous
potential, and which inhibits raising the control electrode
potential for generation of the sub~equent EV. Thi~
coupling can be reduced by incorporating ~till another
electrode to produce a tetrode.
~ 25 A planar tetrode source i5 shown generally at 510 in
¦ Figs. 46-48. A dielectric base 512 features a guide groove
514 in which a planar cathode 516 is located. On the
¦ opposite side of the base 512, and toward the opposite end
thereof, from the cathode 516 is an anode, or counter-
30 electrode, 518. A control electrode 520, ~imilar to the
control electrode 500 shown in Fig. 45, is positioned on the
opposite side of the base 512 from the cathode 546 crossing
under the guide groove 514, and is located between the
longitudinal position of the anode 518 and that of the
35 cathode. Consequently, the control electrode 520 may be
biased and pulsed to effect generation of EV's from the
cathode 516 as described in relation to the triode source
490 in Fig. 45, even wit~h the cathode and anode potentials
~; .
~` 1330827
-80-
held con tant.
A feedback electrode 522 is also positioned on the
~ opposite 9 ide of the base 512 from the cathode 516. The
3 feedback electrode 522 i3 po~itioned sufficiently clo~e to
the anode 518 to diminiqh any coupling between the control
electrode 520 and the anode. Further, as may be appreciated
by reference to Fig. 46, the feedback electrode 522 extends
partly into a rece~s 524 in the side of the anode 518 so
that the anode partially ~hields the feedback electrode from
the control electrode 520 to minimize any inadvertent
coupling between the control electrode and the feedback
electrode.
The tetrode at 510 illustrated in Figs. 46-48 may be
conqtructed utilizing microlithographic film techniqueq.
The width of the EV guide groove 514 may range from
approximately 1 micrometer to approximately 20 micrometerq;
therefore, either optical or electron lithographic methodq
may be u~ed to con~truct the tetrode. Typically, aluminum
oxide may be used to form the dielectric ba~e 512, and
molybdenum may be the conductor material used to form the
various electrodes. Other choice for material~ include
diamond-like carbon for the dielectric and titanium carbide
or graphite for the conductor. In general, any ~table
dielectric material and ~table metallic conductor material
may be utilized. The cathode 516 may be wetted with liquid
metal as diccu~ed hereinbefore. However, with ~mall
structureq in thermal equilibrium, there i~ the po~ible
danger of the migratory metal ~traying to place~ other than
the cathode 516 to alter the electrode configuration.
Alternatively, the planar cathode 516 may be pointed at the
end 526 to provide a sharpened tip to aid in the production
of field emitted electron~ in EV formation, rather than
relying on metal wetting to restore a cathode edge for EV
production. Multielectrode source~ ~uch as the triode 490
and the tetrode 510 illu~trated herein may be operated in
vacuum, or in ~elected gas pres~ure as di~cussed
hereinbefore in relation to other device~
Multielectrode source~ are discus~ed in further detail
,".. , .. , - - - -
`~ 330827 (
-81-
`, in Section 21 on field emi~sion ~ources, wherein an
operating circuit i9 indicated for a tetrode source.
The previously described triode device~, including the
~ separator~, ~electors and launchers, may be provided in
¦ 5 tetrode form a3 well. While several multielectrode
generators are illu~trated and described herein, other
apparatù~ employing two or more electrodes and useful in
Yarious applications and for a range of purpoRe~ may be
adaptable to EV technology. In general, techniques used in
the operation of vacuum can be used effectively in various
~V generation or maDipulation devices.
r,.j
.
~ . 3 0 8 2 7
`` -82-
18._ Electrodeless Sources
Yet another type of EV generator is ~hown generally at
530 in Fig. 49. A generally elongate dielectric envelope
532 features three electrodes 534, 536 and 538, fixed to
exterior ~urfaces of the envelope. The two electrodes 534
and 538 are positioned on opposite ends of the envelope 532
while the intermedlate electrode 536 i9 shown located
approximately one-third of the distance from the electrode
534 to the electrode 538. The end electrode 538 is an
extractor electrode which is used in the manipulation of
EV's after their formation. The remaining electrode~ 534
and 536 are utilized in the formation of EV'~. The
intermediate electrode 536 is in the form of a ring
electrode ~urrounding the envelope 532. In the particular
embodiment illustrated, the ring electrode 536 is located
within the exterior formation of a constriction that defines
an interior aperture 540 separating the interior of the
envelope 532 into a formation chamber 542, to the left as
viewed in Fig. 49, and an exploitation, or working, chamber
544, to the right as viewed in Fig. 49. Likewi~e, the end
electrode 534 is positioned within the depression formed by
an indentation into the end of the envelope 532.
Consequently, the intermediate electrode 536 is
frustoconical, and the end electrode 534 is conical; the
extractor electrode 538 is planar. The indentation and
constriction on which the electrodes 534 and 536,
respectively, are located are not necessary for the
formation of EV's, but serve other purpo~es as discussed
hereinafter. Although the working chamber 544 is
illustrated as approximately twice the length of the
formation chamber 542, the working chamber may be virtually
any length.
When bipolar electrical energy, ~uch as radio frequency
energy, is applied to the first and second electrodes 534
and 536, respectively, mounted on the dielectric envelope
532 which contains a gas, EV's are formed within the
formation chamber 542 even though the external metallic
electrodes are isolated from the internal discharge. A
-` ~3~0827 ~
-83-
cathode is utilized to generate the EV'~ although the
isolated fir~t electrode 534 appear~ as a "virtual
cathode." Such "electrodeless," or isolated cathode, EV
production may be desirable under some conditions, such as
when there is danger of damaging electrodes by ~puttering
action due to high voltage di~charge EV production.
For a given set of parameters such as spacing, gas
pres~ure and voltage, the discharge is particularly
effective in producing and guiding EV's (as discussed in
connection with gas and optical guides, for example), when
the atomic number of the interior ga~ is high. For example,
in the range of effectivenes~, argon ranks low; krypton is
more effective; xenon is the most effective of the three,
assuming the spacing, pre~sure and voltage conditions remain
the same.
Propagation of EV's through the gas within the envelope
532 produces ion streamers, as described hereinbefore,
appearing as very thin, bright lines in the free gas or
attached to the wall o~ the envelope. One or more EV's may
follow along an ion streamer established by an earlier
propagated EV. The fir~t EV of such a series is propagated
without charge balance; subsequent EV'~ passing along the
same ion sheath established by the first EV of the series do
so with charge balance maintained. As multiple EV'~
propagate along the same streamer, the thickness of the ion
sheath increases.
The dielectric envelope 532 may typically be made of
aluminum oxide and have an internal transverse thicknes~ of
approximately 0.25 mm for operation at 3 kilovolt peak
voltage between the two formation electrodes 534 and 536,
with an interior pres~ure of 0.1 atmosphere of xenon ga~.
With such parameters, the spacing between the formation
electrodes 534 and 536 should be approximately 1 mm. The
dielectric may be metallized with silver for the formation
of the electrodes 534-538.
The fru~toconical ~hape of the first electrode 534
tends to stabilize the position of the EV formation. The
annular constriction provides the aperture 540 of
~ '
~ 1 3 3 0 8 2 7
~ --8 4--
r
approximately 5 x 10-2 mm for the remaining above-noted
`~Y parameter~. The aperture 540 permits operation at di~ferent`~ pressures on opposite sides thereof between the formation
chamber 542 and the exploitation ohamber 544, when
appropriate pumping is utilized to produce the pressure
differential by means of gas pressure communication lines
~ (not shown). For example, reduced ga~ pressure in the
I exploitation chamber reduces the guiding effect of the
¦ streamer~ for easier ~elective manipulation of the EV'~.
EV's in the exploitation, or load, chamber may be controlled
by application of appropriately variable amplitude or timing
potentials to the extractor electrode 538, as well as other
¦ external electrodes (not ~hown) for example, for u~eful
manipulation of the EV'~. For a given pumping rate, a
greater pressure differential may be maintained on oppo~ite
~ sides of the aperture 540 for a ~maller diameter aperture.
J The aperture diameter may be reduced to approximately 2.5 x10-2 mm and still allow passage of EV's therethrough. If
the gas pressure in the exploitation chamber is ~uffuciently
low, the EV's will propagate without visible streamer
production as "black" EV's. Furthermore, an electrodeless
source can be con~tructed with a ~maller distance ~eparating
the formation electrodes 534 and 536 whereby EV's can be
I generated with a~ low as a few hundred volts applied.
Moreover, the electrodele~s source may be planar.
~`
~.~, . . . . .
tj '~ 1330827
-85-
19. Traveling Wave Components
One use for EV's generated within a dielectric envelope
such as provided by the source 530 of Fig. 49 is in a
traveling wave circuit, and particularly in a traveling wave
tube. Such a device provides a good coupling technique for
exchanging energy from an EV to a conventional electrical
circuit, for example. In general, an EV current manipulated
by any of the guiding, generating or launching devices
described herein may be coupled for such an exchange of
energy. For example, a traveling wave tube i~ ~hown
generally at 550 in Fig~ 50, and includes a launcher
(generally of the type illustrated in Fig. 25), or cathode,
552 for launching or generating EV's within a cylindrically
symmetric EV guide tube 554, at the opposite end of which is
an anode, or collector electrode, 556. A counterelectrode
ground plane 558 is illustrated exterior to and along the
guide tube 554, and may partially circumscrlbe the guide
tube. The ground plane 558 cannot completely circumscribe
the tube 554 becau9e such construction would shield the
electromagnetic radiation signal from propagating out of the
tube. Appropriate mounting and sealing fittings 560 and 562
are provided for positioning the launcher or cathode 552 and
anode 556, respectively, at the opposite ends of the guide
tube 554.
A conducting wire helix 564 is disposed about the guide
tube 554 and extends generally between, or just overlaps,
the launcher 552 and the anode 556. The helix 564 is
terminated in a load 566, which represents any appropriate
application but which must match the impedance of the helix
to minimize reflections. A pulsed input signal may be fed
to the launcher or cathode 552 through an optional input,
current-limiting, resistor 568. The input re~istor 568 may
be deleted if it con~umes too much power for a given
application. EV energy not expended to the helix 564 is
collected at the anode 556 and a collector resistor 570 to
ground. An output terminal 572 is provided for
communication to an appropriate detector, such as an
oscilloscope, for example" for wave form monitoring.
'.
.
. 133~82~ (
-86-
The velocity of an EV i3 typically 0.1 the velocity of
~; light, or a little greater, and this speed range compares
fa~vorably with the range of delays that can be achieved by
helix and serpentine delay line structures. For example,
the length of the helix 564 and of the EV path from the
launcher or cathode 552 to the anode 556 may be
approximately 30 cm with the helix so constructed to achieve
a delay of approximately 16 ns at a helix lmpedance of
approximately 200 ohm~. The impedance and delay of the
helix 564 are affected, in part, by the capacitive coupling
to the ground plane 558. The lnslde diameter of the glass
or ceramic tubing 554 may be approximately 1 mm or
smaller, with the tubing having an outside diameter of
approximately 3 mm. An EV can be launched at a voltage of
1 kv (determined primarily by the source) at a xenon gas
pressure of 10 torr to achieve an output pulse of
several kv, for example, from the helix 564.
As an example, with a mercury wetted copper wire a3 a
cathode in place of the launcher 552, a xenon gas pressure
of approximately 10-2 torr, an input pulse voltage 600 ns
wide at 1 kv with a firing rate of 100 pulses per sscond
impressed through a 1500 ohm input resistor 568, and with an
anode voltage o~ zero and a target load 570 of 50 ohms, an
output voltage of -2 kv was achieved on a 200 ohm delay line
564 and an output voltage into the target 556 of
-60 volts. A faint purple glow was establi3hed within the
tube 554 and, when a positive input voltage was applied to
the anode 556, visual EV streamers were present for the last
centimeter of the EV run just before striking the anode.
The wave form generated in the helix 564 is a function of
the gas prei3sure. Generally, a sharp negative pulse of
approximately 16 ns in length was produced with the
aforementioned parameters, followed by a flat pulse having a
length that was linearly related to the gas pressure, and
which could be made to vary from virtually zero at preferred
condition3 of minimal gas pressure to as long as one
millisecond. The input pul3e repetition rate may be reduced
for 3uch high gas pressure values to permit clearing of ions
.,~ . .. . .
~,, . , ,. : ., ,
~`l ~`:
:;`1 ' ~ ~
l~Q827
within the tube between pul~e~ to accommodate the long
output pulse. The magnitude of the negative pul~e increased
a~ the ga~ pre3sure decreased. At minimal ga~ presRure,
only a ~harp negative pulse of approximately 16 n~ width wa~
obtained.
A planar traveling wave circuit i~ ~hown generally at
580 in Fig. 51, and may be constructed by lithographic
technology u~ing films of material. A dielectric ba~e 582
include~ a guide channel 584 containing a collector, or
anode, 586. EV'~ are input by a launcher, or other
appropriate device, at the left end of the guide groove 584
a~ viewed in Fig. 51, and are further maintained within the
guide groove by u~e of a counterelectrode (not Vi8 ible) on
the oppo~ite ~ide of the ba~e 582 from the groove.
A ~erpentine conductor 58~ i8 po~itioned on the bottom
~ide of the base 582, underlying the guide groove 414 a~
illu~trated, and ending in a load re~i~tor, or other type
load, 590, as needed. As EV'~ are launched into and guided
down the groove 584, energy of the EV's i~ tran~ferred to
the serpentine conductor 588 and communicated to the load
590. Remaining EV energy i~ absorbed at the anode 586,
which may be connected to a ground re~i~tor, detector or
other load. Although not illu~trated, it i~ preferable to
have a counterelectrode under the ~erpentine conductor,
~eparated by a dielectric layer, to achieve a rea~onable
line impedance and the reduction of radiation and al~o a
dielectric or ~pace layer between the groove and the
serpentine.
As an alternative to placing the conductor 588 on the
bottom of the base 582 oppo~ite to the guide groove 584, the
groove may be covered with a dielectric and a ~erpentine
conductor such a~ 588 placed above the dielectric cover to
overlie the groove. Without ~uch a dielectric cover layer
3eparating the groove 584 from the conductor above, a
counterelectrode mu~t be positioned on the bottom ~ide of
the base 584 under the guide groove to prevent EV'~ from
moving onto the ~erpentine conductor. With such an
arrangement, electron~ emitted during EV propagation down
~. ;
33~827
~ -88-
,
the guide groove 584 may be collected on the serpentine
condu¢tor ~or added energy tran~fer.
Traveling wave tubes or circuit~ as illustrated in
~,' Figs. 50 and 51, for example, thus provide a technique for
converting EV energy into energy that may be communicated by
conventional electrical circuitry. With such techniques,
I electromagnetic radiation from the microwave region to
visible light can be generated by EV pul~es and coupled to
conventional electrical circuitry by ~electively ad~usting
the transmission line parameters and EV generation energy.
~- 20
., : . ,,
.,
30827
20. Pulqe_Generator
An EV is characterized by a large, negative electric
charge concentrated in a small volume and traveling at rela-
tively high ~peed, so that an EV or EV chain can be used to
generate a high voltage faqt rise and Pall pulse. For exam-
ple, any of the device~ described herein for generation of
EV'~ may be utilized in con~unction with a selector, ~uch a~
shown in Fig. 26 or Fig. 27, to obtain the de~ired charge
structure to provide EV' 8 at a capturing electrode whereby
the high charge density of an EV i~ converted to an electro-
magnetic pulse with the desired overall pulse shape. A
~witching, or pulse rise, ~peed as fast a~ approximately
10-14 second~ may be obtained when a 1 micrometer EY bead
containing 1011 elementary charge~ and traveling at O.l the
velocity of light is captured on an electrode sy~tem
designed for the desired bandwidth. The voltage generated
depends upon the impedance of the circuit capturing the
EV's, but will generally be in the range of several kv.
A pulse generator i~ shown generally at 600 in Fig. 52,
and include~ a cylindrically symmetric selector ~hown
generally at 602. A conically-tipped cathode 604, wetted
with conducting material, is po~itioned within a separator
dielectric base 606 and facing an aperture 608 thereof. A
generating anode 610 coats the exterior of the dielectric
base 606, and an extractor electrode 612 is pos itioned a
short distance in front of the base aperture. A generally
cylindrical conducting shield 614 generally circum3cribes
¦ the separator 602, and is clo~ed by a disk 616 of dielectric
material on which i3 mounted the extractor electrode 612. A
conductive metal coating in the ~hape of an annular ring
' provides a conducting terminal 618 on the side of the di~k
¦ 616 facing the shield 614, and makes electrical contact with
the shield. A load resistance 620 provided by a resistor
coating covers the annular surface area between the
extractor electrode 612 and the ring conductor 618 so that
the separator 602 i9 nearly completely surrounded by
~hielding to limit electrical ~tray fields and to help
complete current paths with minimal inductance. The overall
c~
:."':". ' :
. . . . -~
,
.....
^. . ~ . , , ~ .
i ~ ~330827
go
size of the pulse generator may be approximately 0.5 cm.
The external side of the dielectric disk 616, shown
~ also in Fig. 53, is virtually a mirror image of the interior
i side, featuring a circular output electrode 622 connected to
an annular ring electrode 624 by a resi~tive coating 626,
with the shape and dimensions of the exterior electrodes 622
and 624 being e3sentially the same as those of the interior
electrodes 612 and 618, re~pectively. The output electrode
622 is thu~ capacitively coupled to the extractor electrode
612 whereby the capture of the relatively high charge of an
EV or EV chain by the extractor electrode produces a
corresponding high negatlve charge on the output electrode.
To initiate EV production, an appropriate negative
pulse may be applied to the cathode 604 by mean~ of an input
terminal 628 with the anode 610 maintained at ground, or a
relatively small positive potential, by means of a terminal
630 passing through an appropriate opening 632 in the ~hield
614. A more positive extractor voltage is applied to the
extractor electrode 612 through a terminal 634 to the shield
614 connected to the extractor electrode by means of the
conducting ring 618 and the internal resistor coating 620.
When an EV is generated and leaves the selector 602, and i9
captured by the extractor electrode 612, the potential of
the extractor electrode is rapidly lowered, and rises as the
EV charge is dispersed by means of the resistor coating 620
and the shield 614, and ultimately by way of the terminal
634. The extractor voltage applied to the extractor
electrode 612 is variable 90 that only selected EV's may be
extracted from the selector 602 to provide the output pulse~
as de~ired. A bias voltage may be placed on the output
electrode 622 by a terminal 636 connected to the ring
oonductor 624 and ulti~ately to the output electrode by the
resistor coaking 626.
In general, for fast pulse times, small, low reactance
components with a minimum distance between the various
circuit elements are used. The approach distance of the EV
from the selector 602 to the extractor electrode 612, and
the charge of the EV determine the ri~e time of the negative
i
:~
~:'` ~ `
1330827
9 1
pulse on the output electrode 622. The RC constant, or
resistance, of the load resistor 620 determines the pulqe
fall time. For example, output pulses with a ri~e and fall
time of 10-13 secondq minimum may be achieved with the
"picopulser" 600 having a maximum external diameter of
approximately 0.5 cm. The load resistor 620 is typically at
least aq large as about 10-4 ohms (and can be 10-3 ohms),
and may be achieved by utilizing a thin metallic coating on
the surface of the dielectric disk 616, which may be
ceramic, for example. A similar resistive coating may be
used as the resistor 626 to achieve the output coupling and
bypass capacitor action. The output resistor 626 determine~
the bias on loads, for example. Where dc current is drawn
at the output, the output pulse decay times may be varied by
varying the output resistive coating 626, with longer pulse
decay times achieved by increasing the resistance value of
the coating, utilizing fired-on thick film fabrication
techniques, for example. An operating voltage of up to 8 kv
for various biases can be obtained, with proper attention to
the finish of the metal conductive coating rings 618 and
624. The level of the output pulse may be varied by
selectively varying the attenuation factor in the load
circuit applied to the termi-nal 636.
The picopulser 600 thus provides a technique for
achieving very fast and large voltage pulses by initial
generation of EV's or EV chains. For optimum performance,
the pulse generator 600 should be operated in vacuum.
-~
,
--~ 1330827 `
-92-
21. Field Emission Sources
_ _ . __
The principle requirement for generating an EV is to
rapidly concentrate a very high, uncompensated electronic
charge in a ~mall volume. Such an operation implies an
emission process coupled to a fast switching process. In
the various gaseous EV generators described hereinbefore,
the ~witching proceqs is provided by non-linear actions of
3 ga~ ionization and pos~ibly some electronic ram effects.
The gas ~witching process operates even with the sources
utilizing cathodes wetted with liquid metal, once the basic
field emission process liberates metal vapor from the
cathode region by thermal evaporation and ionic
~ bombardment. Pure field emi~sion generation of EV's can be
Z achieved with the elimination of all gas and migratory
Zl 15 material from the system of EV generation. To achieve such
'd field emission generation, fast switching must be provided
and coupled to the field emitter so that the emi~ion
Zl process can be switched on and then off again before the
emitter is heated to the evaporation point by electronic
conduction. Thus, EV'q are generated by a field emission
Z cathode operated in the emission density region beyond that
normally used with other field emission devices, by pulZing
the emitter on and off very rapidly, that is, faster than
the thermal time constant of the cathode9 thereby preventing
Z 25 thermal destruction of the emitter. Since the thermal time
; conZtant of the emitter is typically less than 1 picosecond,
the resulting required short Zwitching time for potentials
in the hundreds of volts range can be achieved using EV-
actuated switching devices, such as the pulse generator 600
illustrated in Figs. 52 and 53~
A field emission EY source is shown generally at 650 in
Fig. 54~ and is constructed and ~unctions similarly to the
pulse generator 600 of Figsr 52 and 53 with the exception
that the pulse output electrode 652 of the field emission
source includes a pointed emitter 654 extending from the
otherwise diqk-shaped electrode. An appropriate voltage
pulse signal is applied to the cathode 656 and anode 658 of
the separator ~hown generally at 660 to generate EV'~, and a
F'=' ' ' . ,
, ~:
rD: ~ ~
~ ~330827
-93-
selected extractor voltage is applied to the extractor
electrode 662 to attract an EV thereto. Capture of the EV
at the extractor electrode 662 produces a fa~t rise negatlve
pulse on the output electrode 652 so that a large field i~
concentrated at the tip of the emitter 654. The re~ulting
field effect at the tip of the emitter 654 produce~ one or
more EV'~ by pure f~eld emi~3ion, with the f$eld emi~ion
source operating in vacuum~ The EV-generated negative pul~e
on the output electrode 652 must also have a ~hort fall time
0 90 that the pulse i~ killed before the emitter 654 is
damaged in the decline of the pulse. The resistor coating
664 on the extractor electrode side of the di~k 666 may be
approximately 10-2 ohms, and the resistor coating 660 on the
field emitter side may be approximately 106 ohms. An EV
guide, 670, of the generally cylindrical construction
illustrated in Fig. 15, for example) is shown positioned to
receive EV's launched from the emitter S54 and to manipulate
them to whatever load i~ intended.
The field emis~ion generator 650 may be used to form
EV'~ while at the same time testing the field emis~ion
cathode 654 for damage in order to optimize the formation
process to minimize damage. A pho~phor screen, or a witness
plate (not shown), may be positioned appropriately to
receive EV's formed at the emitter 654. The picopul~er is
turned off and a bia~ voltage is applied through the lead
672 to impress a dc voltage on the emitter 654 to draw dc
field emission therefrom. Although the bia~ voltage applied
to the lead 672 is usually negative, it can be positive if
the EV from cathode 656 is produced by a voltage higher than
2 kv. Then, the emission pattern on the ad~acent pho~phor
screen or witnes~ plate may be analyzed in conjunction with
the value of the dc bias voltage and current to the emitter
554 to determine the cathode radius, crystallographic ~tatus
and other morphological characteristic~ immediately after EV
generation. Such analysis methods for field emi~sion
surface~ are well known.
The peak voltage of the picopulser being u~ed to drive
the field emitter 654 can be determined by varying the bias
.'~
~`".,. : -
1 3 3 0 8 2 7
-94-
voltage through the lead 672 to offset the puli~e voltage to
the cathode 656. In this way, the field emitter 654 is
being used as a very high speed rectifier or detector to
measure the pulse peak to the cathode 654. To test
d 5 characteristics of the EV's produced, a film or foil of
~mooth metal, as a witne~s plate, may be positioned in front
d 0~ an anode (not qhown) positioned in front o~ the emitter
654, and connected to that anode. A spacing of up to one
j millimeter between the emitter 654 and such an anode can be
used in vacuum when the Rystem i3 operated at approximately
2 kv. The impact mark the EV leaves on the wltness plate
can be analyzed in a scanning electron microscope to
determine the number of EV bead~ formed and their pattern of
arrival. Many high speed effects can be invesitigated with
the generator 650 of Fig. 54. If the output from the pulse
generator i~ kept low in voltage and a sensitive detector
used for detecting emission from the field emitter 654, it
is possible to effectively measure very ~hort pulse voltage
amplitude by a substitution technique using the high speed
rectification ability of the field emitter. The bias
voltage applied through lead 672 is substituted for the
pulse voltage.
At high levels of pulse voltage, far into what is
u3ually thought of as the space charge ~aturation region for
a field emitter, the emitter 654 generates bunches of
electron~ that reRemble EV~SJ as detected on a nearby
witness plate. These ~mall EV~S are potentially very useful
for specialized computer-like applications using charge
steering.
The field emiqsion generator 650 shown in Fig. 54 is an
example of one of the vays relatively large components can
be utilized in reaching the necessary switching speeds to
achieve pure field emission EV production. For practical
application, it may be desirable to use a complete Rystem of
compatible microcomponents to fabricate the switching and
launching devices. Moreover, in view of the small sizes and
relatively high voltages required, more practical devices
for utilizing and generating EV's formed from relatively
.~ :
.
13~827
-95-
pure field emis~ion may be con~tructed utilizing microfabri-
cation.
Fig. 55 shows a microcircuit using thin film
techniques to construct a complete system for producing EV's
by field emlssion without relying upon external EV
generators or bulk components that might complicate high
~peed operation. Here, the switching process is carried out
by feedback on a time scale consistent with the thermal
processes in the EV generator, that is, the switching rate
is equal to or, preferably, faster than the thermal time
constants and thermal processe~. It is necessary to switch
the emitter on and off in le~s than 1 p~ to prevent cathode
destruction.
The field emission source shown generally at 680 in
~ 15 Fig. 55 is similar in construction to the tetrode source
j 510 of Figs. 46-48. Thus, a dielectric ba3e 682 features an
elongate groove 684, which may be of generally rectangular
cross section, in which is positioned a line cathode source
686 which is operated ~ithout being wetted with a metallic
coating. A counterelectrode 688 is pos itioned on the
¦ opposite side of the base 682 from the groove 684 and
toward the opposite end of the base from the cathode 686.
The counterelectrode 688 underlies a portion of the guide
groove 684. A control electrode 690 is also positioned on
the same side of the base 682 as the counterelectrode 688,
and extends from a side edge of the base to a position
underlying and crossing under the guide groove 684 between
-the ends of the cathode 686 and the counterelectrode. A
feedback electrode 692 is also positioned on the opposite
side of the base 682 from the cathode 686, and extends
laterally across the underside of the base toward the end of
the counterelectrode 688 closer to the cathode. A leg 694
of the feedback electrode 692 extends along a rece~ 696 in
the counterelectrode 688 whereby the feedback electrode may
lnteract with a generated EV during the propagation of the
EV along the guide groove 684, generally for the length of
the electrode leg 694.
~,. .~,
,~
';~; ' ' ' . - :
.
!`'
/
`~ ~ 1330~2~
-96-
Fig. 56 ~hows a circuit diagram at 700 of the field
;l emis~ion ~ource 680 of Fig. 55 and associated apparatus for
effecting the field emi~ion production of EV'~. An energy
storage device 702 is connected to the cathode 686, and
provided with appropriate negative potential through a lead
¦ 704. The pas~ive energy ~ource 702 may be a capacitor or a
¦ strip delay line, as used in hydrogen thyratron pul~e radar
system~ for example, with a resi~tor or conductor feed. The
generating energy source 702 typically provide~ a 1 ps
negative pulse when discharged by mean~ of the potential
change on the control electrode 690. Otherwise, a constant
potential may be applied between the cathode 686 and the
counterelectrode 688.
A phase inverting air core pulse transformer 706 is
selectively operated by a trigger pulse through a lead 708
to apply a po~itive control bias voltage, supplied by means
of a lead 710, to the control electrode 690 to initiate the
EV field emi~sion generation at the cathode 686. The
feedback ~ignal needed to 3ustain emi~sion after the trigger
pulqe has been removed, and until the stored energy in the
power supply 702 has been depleted, i8 provided by the
tran~former 706 by means of the feedback electrode 692.
The field emitters, such as 654 and 686, used in pure
field emission sources such as those described, should be
fabricated from relatively ~table material in terms of
thermal and ion sputter damage. For example, metal
carbide~, such as titanium carbide and graphite, provide
such characteri~tics to make good cathodes. Similarly, the
dielectric material should be of high stability and high
dielectric field strength. Aluminum oxide and diamond-like
carbon filmq exhibit such characteri~tics. Since there is
no self-repairing process available for the cathodes, as
there is with liquid metal wetted source~, it is preferred
to use ultra high vacuum at the emitters to avoid damage
thereto by ion bombardment, or modification of the surface
work function.
Prevailing factors preclude the use of pure field
emitters of large size. The critical limit appears to be
s, ~-
s~
~: - .
-~` 1330827 ~
$ -97-
¦ approximately one micrometer for the lateral dimen~ion of an
emitter of the type 686 ~hown in Fig. 55. For cathode~
above ~uch ~ize, the ~tored energy of the as~ociated
f., circuitry places an undue thermal ~train on the ~mall
emitter area during emi~ion. Below the one micrometer ~ize
range, the field emitter ha~ the advantage of large cooling
effect~ provided ~mall elements having a naturally high
surface-to-volume ratio.
, ~
~: ,
1330827`
98 -
22. X-Ray Source
EV's may be utilized to generate X-rays. An X-ray
generator, or source, is shown generally at 720 in Fig. 57,
and include~ a mercury wetted copper type cathode 722, as
illustrated in Fig. 4, and a separator 724 equipped with a
counterelectrode 726, as shown ln Fig. 8, positioned
relative to an anode 728 for generation and propagation of
EV's, including pos3ibly EV chains, from the cathode through
;~ the separator aperture to the anode.
10It has been found that stoppage of an EV on a material
target or anode is accompanied by a flash of light from the
plasma produced and a crater left as a result of the
disruption of the EV and accompanying expenditure of
energy. A portion of the energy expended is carried off in
lSX-ray production. The X-ray source itself within the target
728 is as small as the EV, that is, in a range of
approximately 1 to 20 micrometers in lateral dimension,
depending upon how the EV was originally made or selected.
The small source of X-rays has a relatively high production
efficiency and intensity, providing a high total X-ray
output compared to the input energy. This phenomenon
indicates an intense X-ray production upon disruption of the
ordered EV structure, pos3ibly due to the sudden disruption
of the large magnetic field generated by electron motion
within the EV.
Output from the cathode 722 and separator 724 impinges
on the anode target 72~ to produce emi~sion of X-rays as
indicated schematically in Fig. 57. The material of the
target 728 is sufficiently low in inductance to cause the EV
to effectively break apart. A low atomic num~Rr material,
~uch as graphite, minimizes damage due to EV disruption, and
allows relatively ea~y passage of X-rays produced to the
output side of the target 728. The X-ray source 720 can be
operated either in vacuum or in ~ low pressure gas. For
¦ 35 example, in an environnent of a few torr of xenon gas, the
cathode 722 and separator 724 may be spaced as far as
approximately 60 cm fro~ the anode target 728, with a pulse
signal of 2 kv applied to the cathode for the production of
~. ,- .
' ` 1330827
_99_
EV'~. Analy~is of the total X-ray output from the ~ource
1 720 can be accompli~hed utilizing known techniques, ~uch a~
u~ing filter~, or photographic film, or wavelength
di~per~ion ~pectrometers. However, ~ince the X-ray photons
are all generated at approximately the same time, energy
I di~persive ~pectrometer~ are not able to analyze the
~pectral energy content of the X-ray output.
I The present invention thus provide~ an X-ray generator,
or ~ource, capable for u~e a~ a point source of X-ray~ for
application in ~top motion X-ray photography, for example.
The X-ray generator of the pre~ent invention can
additionally be u~ed in a wide range of X-ray applications.
`~
3 3 0 g 2 7
,~,,, -100-
. .
23. Electron Source
EV's moving along a guide will generally produce the
emission of electrons, which may be collected by a collector
electrode, for example. In the case of RC guides, for
example, it is possible to collect electron emission out the
top of the guide groove if the groove is sufficiently deep
and the EV is strongly locked to the bottom of the guide
groove, or at least the counterelectrode on the opposite
~ide of the dielectric base of the guide. The electrons
thus emitted come from 3econdary and field emission sources
that have been produced by the energy o~ the passing EV.
Since these electrons have come from a dielectric material
with a relatively long RC time constant for recharge, it is
necessary to wait for such recharge until another EV can
occupy the region, and thereby cause further electron
emission. In the LC class of guides, this time delay is
relatively short since recharge is supplied by wa~ of
metallic electrodes. Electrons can be collected for dc
output use by simply supplying a collector electrode, since
the emitted electrons have been given initial energy by the
EV. In the case of LC guides, any of the electrodes in the
guide structures of Figs. 20 or 21 can be utilized as
collector electrodes.
The characteristic of an EV that it can cause electron
emission enables the EV to be effectively used as a cathode
for various applications. A properly stimulated EV can be
made to emit a fairly narrow band of electron energies. The
primary consideration in using this type of cathode is
determining the mean energy and the energy spread of the
emitted electrons. There is also a chopping effect that
results from having a definite spacing between the EV's
moving along a guide and producing electron emission, for
example. The chopping range is generally available from
essentially steady emission from a virtually continuous
train of EV'~ to a very pulse-like emission from passing a
~ingle EV or EV chain under an aperture. Consequently, the
nature of the EV propagation as well as the guide structure
through which the EV's are moving must be chosen approprlate
,,
`
,
. '
~ ~330827~
101-
to the application of the electron emission.
A gated, or chopped, ele¢tron emis~ion sour¢e is shown
generally at 740 in Fig. 58, and may be part of a trlode-
like structure. An RC EV guide 742 is provided, featuring a
guide groove 744 and a counterelectrode (not visible) on the
underside of the guide ba3e from the groove generally like
the EV guide illustrated ~n Fig. ll. A dielectric plate 746
iq po~itioned immediately over the ba~e of the guide 742.
The plate 746 features openingq 748 which overlie the guide
grooYe 744~ and are lined with metal coatings 750 which
~erve as gating electrode~. A third element, not shown~ may
be an anode or the like po~itioned above the dielectric
plate 746 to receive or ~ollect the emitted electrons; the
exact nature of the third element is dictated by the use to
which the electron emission i9 to be applied.
In operation, one or more EV'q are launched or
otherwi~e propagated into the guide groove 744 as indicated
by the arrow I. As di~cussed hereinabove, qecondary or
field emission effects a~sociated with the pas~age of the EV
down the guide groove 744 result in electron emi~ion which
may be propagated out of the guide groove, as indicated by
the arrow J, the electron~ having been given initial
propagation energy in their formation associated with the
presence of the EV. In general, the emitted electronq may
be further attracted by the third component, ~uch as an
anode (not show~). However, electron propagation to the
third component is selectively controlled by the application
of appropriate potentials to the control electrodes 750. In
general, the potential applied to a control electrode 750
will alwayq be negative relative to the cathode used to
generzte the EV's. In effect, the gate, or opening, 748 in
the dielectric 746, in each case, may be opened or closed to
electron passage therethrough by ~e'ection of the specific
potential on the re~pective control electrode 750. To close
the gate 748, the potential on the control electrode 750 is
made more negative so that no electron emission will take
¦ place therethrough. To open the gate 748, the potential on
the control electrode 750 i~ made lesq negative, that is,
,.... . .. , ~
~330827 ~
-102-
.:
relative to the EV-generating cathode, and electron emis~ion
through the gate i~ permitted.
As an EV propagate~ down the guide groove 744, the
electron emission i9 generated. However, electrons may pas~
through the dielectric plate 746 to the third electrode
component only at the locations of the passageways 748.
Consequently, an EV moving along the guide groove 744 causes
electron pulses to be emitted through the dielectric plate
746, with the pul~es occurring at the location3 of the
pa88age8 748. Further, a given pas~age 748 may be closed to
electron transmission therethrough by the appropriate
potential being placed on the re~pective control electrode
750. Consequently, a selective pattern Or electron emission
pulses may be achieved by appropriate application of
potentials to the control electrodes 750. The pulse pattern
may be further varied by propagating a train of EV's or EV
chains down the guide groove 744 to achieve, for example, an
extended pattern of electron emi~sion pulses along the array
of ports 748, with the potential values placed on the
various control electrodes 750 themselves changing with
time. Consequently, the electron emisqion pattern may be
varied extensively by both the ~election of EV propagation
as well as the modulation of potentials on the control
electrodes 750.
To prevent the EV itself from exiting one of the ports
748, the groove 744 should be maintained relatively deep, or
alternatively, a spacer (not shown) can be u~ed between the
plate 746 and the base of the guide 742.
It will be appreciated that a pattern of electron
emis~ion ports 748 ~ay be provided as desired, with
appropriate EV guide mechanisms positioned in con~unction
therewith. The number and positioning of the ports 748
along the guide groove 744 may be varied to 3elect the
electron emission patters~ as well. The electron emission
ports 748 may also be effectively throughbores in a
dielectric plate which completely circumscribes each port,
for example. In such ca~e, the control electrodes 750 may
also line the port wall~ on all sideq.
~ -~ 133~827 ~
:f~ -103-
ff~
";,
In general, any type of EV generator that produce~ the
de~ired EV output for the given application may be utilized
to provide the EV's for electron emis~ion. Typically, a
verqion of the electrodele~ ~ource illustrated in Fig. 49,
~ 5 operating at a low gaq pre~ure, may be utilized. The inert
¦ ga~ pre~ure in the ~ystem might be in the range of 10-3torr, and would be in equilibrium throughout the ~y3tem.
Electron emis~ion by EV propagation, utilizing any of
the apparatu~ de~cribed herein, ~uch a~ the gated electron
~ource 740 illu~trated in Fig. 58, may find variou~
applications. For example, various devices until now
impractical for failure of the prior art to provide a
cathode of ~ufficient emls~ion inten3ity may now be
exploited using an EV-generated electron ~ource such a~
di~closed herein. Such a cla~s of device~ a~ the beamed
deflection, free electron device, for example, may be
provided utilizing a gated electron ~ource of the type
illustrated in Fig. 58, for example.
~f
1 35
.,.. , - . ~
~ ~ .
::
~. . , -
. . . .
1330827
- 104 -
24. RF Source
Pa~sage of EV's through the LC guides of Figs. 20 and
( 21 generates RF fields within the guides, but the interaction
with such fields is utilized to guide the EV's, and not to
exploit external radiation. However, RF generated by
pas~age of an EY can be coupled out of an EY guide and made
available for external application.
¦ Fig. 59 illu~trates a general form of an RF source, or
generator, ~hown generally at 760. A dielectric base 762
featuring an elongate guide groove 764 provides a guide
structure for EV's entering the groove, as indlcated by the
arrow K. A counterelectrode 766, which may be po~itioned on
the underside of the dielectric base 762, feature~ a series
of slot3 7~8. The RF production involves a charge induced
field on the counterelectrode 766. The results are inten~e
if the counterelectrode is in slotted form. A second
electrode, in the for~ of a collector, 770 is positioned
below the counterelectrode 766, and separated therefrom by a
dielectric. This latter dielectric may be space, or a layer
of dielectric material (not shown). The collector 770
features a ~eries of arms, or extensions, 772, with one such
extension positioned directly below each of the
counterelectrode slots 768. As EV's move along the guide
channel 764, the counterelectrode slots 768 provide openings
for the charge of the EV's to couple to the collector 770
wherein the RF field is produced. The RF energy can be
tapped from the collector 770 by any appropriate circuit, or
further radiation system.
There is a reciprocal relationship between the EV
velocity along the guide channel 764 and the output cavities
768, in con~unction with the collector electrode arms 772,
that determines the frequency of the radiation provided.
The frequency produced i~ equal to the ~peed of the EV
multiplied by the inver~e of the ~pacing between the slot~
768.
It will be appreciated that the shapes of the opening~
768 in the counterelectro~e 766 determine the waveforms to
be produced. Aperiodic wave forms, which may be employed for
, - . ,
,.'`~` ~
1 33 0827 i:
-105-
drivlng variou~ computer or timing function~, can be
generated with the structure shown in Fig. 59 by
( approprlately shap1ng the counterelectrode o~enings 768.
The load on the collector eleotrode 770 must be
proportloned accordlng to the bandwidth of the generated
waveform. For low frequencies, the output of the collector
electrode 770 should be connected to a transmis~ion line
wlth re31stive termination at it~ characteristlc
impedance. The velocity of the EV's in the guide groove 764
can be locked into synchronou3 motion by using RF inJection
or interaction as noted hereinbefore in the discussion of LC
guides. Such synchronization helps regulate the periodic
rate of the output pulses obtained from the collector
electrode 770.
The wave form generator of Fig. 59 can be operated to
provide either positive or negative polarity pulses by
differentiation of the EV charge as the EV passes the slot
768 in the counterelectrode 766. A high impedance load on
the output of the collector electrode 770 produces
es3entially negative pul~e~. However, a low load impedance
on the counterelectrode 770 result~ in the production of
fir~t a negative pulse and then a positive one. Thi~ pulse
form is useful for generating po~itive wave forms used in
driving field emi3~ion devices into the emitting state, as
an example of but one application o~ the use of EV'~ to
generate electromagnetic energy.
;
.- ... . . . . .
`'!~;,.` ~ :
,:
- ' 1330827
~ -106-
.
25. Conclu~ion
The pre~ent invention provides techniques for genera-
( ting, isolating, manipulating and exploiting EV's, either a~
~ndividual EV beads or as EV chainq. Control of generation
and propagation of EV's has extenqive applications, some of
which have been noted hereinbefore. The propagating EV's
themselve3 are sources of energy, 1ncluding electromagnetic
energy in the RF range available by utilizing an EV ~F
~ource, such a~ illustrated in Fig. 59, or a traveling wave
device, ~uch as illustrated in Fig. 50 or 51. The emission
of electrons accompanying EV propagation across a dielectric
surface, for example, enables the propagating EV's to be
treated as a virtual cathode with the use of the EV source
of Fig. 5~, for example. By appropriate selection of the
gating pattern in such an electron source, a variety of
applications are available wherever intense electron beams
are required, for example. The picoscope described herein-
before also utilizes electron emis~ion attendant to EV
propagation to provide a fast response, miniature oscillo-
i 20 scope for analysis of electrical signalq, for example.
Similarly, the picopulser of Fig. 52 utilizes the rapid
communication of large electric charge to produce fast rise
and fall high voltage pulses. Such fast pulses have a
variety of u3es, including the operation of pure field
emission device~, such as the EV generator of Fig. 54.
The ability to produce and ~electively manipulate EV'~
provides a new electrical technology with several very
desirable feature~. In general, the components of this
technology are extremely small, and operable over a range of
applied voltage. As noted, operations carried out with the
EV technology are very rapid, and involve the rapid
communication of large concentrations of energy in the form
of the EV's. The various generators, launchers, guides,
separators, ~electors and splitters, for example, are
analogous to vacuum tubes, transi~tor~ and the like of prior
art electronic technology, for example.
It will be appreciated from the foregoing di~closure of
the present invention that the various deviceq described
.FF,~
,,~
,
133~827
1 -107-
,1
herein may be combined to fit given application~. h
generator from the various generators diqclosed herein may
be utilized with one or more guide devices to provide the
EV's utillzed in a picoscope, for example. An EV generator
5 may be comblned with guldes and one or more qplitters and/or
one or more switches to provlde multiple EV pathq which, in
the ca_e of the qwitche~, may be selected for EV
propagation. An EV generator may be combined with guides
and one or more picopulsers to provide pulqe output~ at
deqired locations and, utilizing a variable time delay arm
of a splitter such aq illustrated in Fig. 33, to provide
time variable pul_ing. Similarly, any of the energy
¦ conversion device~, quch a_ the traveling wave circuits of
Figs. 50 and 51, or the RF source of Fig. 59 or the electron
emiqsion source of Fig. 58, may be combined with the variou_
other EV manipulation components such a_ guides, spl~tters
and switches. It will further be appreciated that EV
selectors, separators and launchers may be utilized where
appropriate to provide EV's of the desired charge _ize,
launched into a specified guide or other device, and free of
plasma discharge contaminantq. The electron camera itqelf
is usa~le in analyses of EV behavior itself, as well as in
other analy~es, including but not limited to the analy~es of
time-varying electric fields through the combination with
the picoscope, or the multi-dimen~ional scope arrays
illustrated in Fig. 44, for example.
The foregoing disclosure and description of the
invention i~ illustrative and explanatory thereof, and
various changes in the method step~ as well as in the
details of the illustrated apparatus may be made within the
scope of the appended claims without departing from the
spirit of the invention.
