Low-temperature electrical discharge through solid xenon
The uniform self-sustained electrical discharge through solid xenon has been realized and studied. The multiplication of electrons proceeds in the noble gas above the xenon crystal interface whereas a positive feedback is realized at the account of multiple exciton formation by excess electrons dr...
Saved in:
| Date: | 2008 |
|---|---|
| Main Authors: | , , , |
| Format: | Article |
| Language: | English |
| Published: |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
2008
|
| Series: | Физика низких температур |
| Subjects: | |
| Online Access: | https://nasplib.isofts.kiev.ua/handle/123456789/117884 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Cite this: | Low-temperature electrical discharge through solid xenon / E.B. Gordon, V.I. Matyushenko, V.D. Sizov, B.M. Smirnov // Физика низких температур. — 2008. — Т. 34, № 11. — С. 1203-1211. — Бібліогр.: 28 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
nasplib_isofts_kiev_ua-123456789-117884 |
|---|---|
| record_format |
dspace |
| spelling |
nasplib_isofts_kiev_ua-123456789-1178842025-02-09T21:25:10Z Low-temperature electrical discharge through solid xenon Gordon, E.B. Matyushenko, V.I. Sizov, V.D. Smirnov, B.M. Физические свойства криокристаллов The uniform self-sustained electrical discharge through solid xenon has been realized and studied. The multiplication of electrons proceeds in the noble gas above the xenon crystal interface whereas a positive feedback is realized at the account of multiple exciton formation by excess electrons drifted through the crystal: molecular excitons emit VUV photons which knocked out secondary electrons from photosensitive cathode. The discharge was stimulated by short electrical spark along the sample axes. The discharge electrical properties as well as the spectra of solid xenon electroluminescence in UV and visible have been studied. Electric discharge in solid xenon was proved to be an effective source of UV radiation and a convenient tool to study the processes involving excitons and electrons in solid xenon at high pressures. This work is supported partially by Russian Foundation for Basic Researches grant 07-03-00393. 2008 Article Low-temperature electrical discharge through solid xenon / E.B. Gordon, V.I. Matyushenko, V.D. Sizov, B.M. Smirnov // Физика низких температур. — 2008. — Т. 34, № 11. — С. 1203-1211. — Бібліогр.: 28 назв. — англ. 0132-6414 PACS: 64.70.kt;52.80.Wq;52.80.Yr;79.20.Kz https://nasplib.isofts.kiev.ua/handle/123456789/117884 en Физика низких температур application/pdf Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| topic |
Физические свойства криокристаллов Физические свойства криокристаллов |
| spellingShingle |
Физические свойства криокристаллов Физические свойства криокристаллов Gordon, E.B. Matyushenko, V.I. Sizov, V.D. Smirnov, B.M. Low-temperature electrical discharge through solid xenon Физика низких температур |
| description |
The uniform self-sustained electrical discharge through solid xenon has been realized and studied. The
multiplication of electrons proceeds in the noble gas above the xenon crystal interface whereas a positive
feedback is realized at the account of multiple exciton formation by excess electrons drifted through the
crystal: molecular excitons emit VUV photons which knocked out secondary electrons from photosensitive
cathode. The discharge was stimulated by short electrical spark along the sample axes. The discharge electrical
properties as well as the spectra of solid xenon electroluminescence in UV and visible have been studied.
Electric discharge in solid xenon was proved to be an effective source of UV radiation and a convenient tool
to study the processes involving excitons and electrons in solid xenon at high pressures. |
| format |
Article |
| author |
Gordon, E.B. Matyushenko, V.I. Sizov, V.D. Smirnov, B.M. |
| author_facet |
Gordon, E.B. Matyushenko, V.I. Sizov, V.D. Smirnov, B.M. |
| author_sort |
Gordon, E.B. |
| title |
Low-temperature electrical discharge through solid xenon |
| title_short |
Low-temperature electrical discharge through solid xenon |
| title_full |
Low-temperature electrical discharge through solid xenon |
| title_fullStr |
Low-temperature electrical discharge through solid xenon |
| title_full_unstemmed |
Low-temperature electrical discharge through solid xenon |
| title_sort |
low-temperature electrical discharge through solid xenon |
| publisher |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| publishDate |
2008 |
| topic_facet |
Физические свойства криокристаллов |
| url |
https://nasplib.isofts.kiev.ua/handle/123456789/117884 |
| citation_txt |
Low-temperature electrical discharge through solid xenon / E.B. Gordon, V.I. Matyushenko, V.D. Sizov, B.M. Smirnov // Физика низких температур. — 2008. — Т. 34, № 11. — С. 1203-1211. — Бібліогр.: 28 назв. — англ. |
| series |
Физика низких температур |
| work_keys_str_mv |
AT gordoneb lowtemperatureelectricaldischargethroughsolidxenon AT matyushenkovi lowtemperatureelectricaldischargethroughsolidxenon AT sizovvd lowtemperatureelectricaldischargethroughsolidxenon AT smirnovbm lowtemperatureelectricaldischargethroughsolidxenon |
| first_indexed |
2025-11-30T23:06:06Z |
| last_indexed |
2025-11-30T23:06:06Z |
| _version_ |
1850258457152716800 |
| fulltext |
Fizika Nizkikh Temperatur, 2008, v. 34, No. 11, p. 1203–1211
Low-temperature electrical discharge through solid xenon
E.B. Gordon
Institute of Problems of Chemical Physics RAS, Chernogolovka 142432, Russia
E-mail: gordon.eb@gmail.com
V.I. Matyushenko and V.D.Sizov
Institute of Energy Problems of Chemical Physics RAS, Chernogolovka 142432, Russia
B.M. Smirnov
Joint Institute for High Temperatures RAS, Izhorskaya 13/19, Moscow 127412, Russia
Received February 15, 2008, revised May 21, 2008
The uniform self-sustained electrical discharge through solid xenon has been realized and studied. The
multiplication of electrons proceeds in the noble gas above the xenon crystal interface whereas a positive
feedback is realized at the account of multiple exciton formation by excess electrons drifted through the
crystal: molecular excitons emit VUV photons which knocked out secondary electrons from photosensitive
cathode. The discharge was stimulated by short electrical spark along the sample axes. The discharge electri-
cal properties as well as the spectra of solid xenon electroluminescence in UV and visible have been studied.
Electric discharge in solid xenon was proved to be an effective source of UV radiation and a convenient tool
to study the processes involving excitons and electrons in solid xenon at high pressures.
PACS: 64.70.kt Molecular crystals;
52.80.Wq Discharge in liquids and solids;
52.80.Yr Discharges for spectral sources;
79.20.Kz Other electron-impact emission phenomena.
Keywords: solid xenon, electrical discharge, exciton emission.
1. Introduction
Electric discharge requires for its self-maintaining the
reproduction of electrons and ions which go to electrodes
[1]. In gases this reproduction results from atom ioniza-
tion by electron impact. In contrast to gases where the
energy distribution function of electrons has a long tail,
in condensed heavy gases due to effective excitation of
atoms the ionization of an atom occured by impact with
more fast electrons is absent. Nevertheless, the conditions
of self-maintaining may be satisfied in solid inert gases if
electrons are injected from outside where they are formed
under the cathode bombardment by VUV photons which
in turn are generated inside the condensed inert gas by
these electrons drifted in an external electric field. This
scheme was formulated [2,3] and then was shown to be
realized in [4,5] for the solid xenon. The aim of present
paper is to create the reproducible uniform electric dis-
charge in solid xenon by brute force technique applicable
in principle for solid xenon at high pressures. The brief
results of the studying such a discharge through solid
xenon at ambient pressure are reported.
2. Peculiarities of electron drift in solid xenon
The mean free path of an excess electron in liquid and
solid heavy rare gases amounts to many meters [6,7] and
electron mobility exceeds that in a coin metal or in a gas
of atmospheric pressure [8]. Meanwhile the absence of
own free electrons makes possible there, unlike in metals
and semiconductors, sustaining the electrical field inside
a sample. Naturally the question appears [2,3], is it possi-
ble to organize an electrical discharge in the solid to ex-
cite and to ionize a matrix as it takes place in a gas? The
answer depends first of all on the fact to what extent the
unique high mobility of electron and its unique low elas-
tic losses save themselves at high electron energy, be-
cause the growth of electron kinetic energy should result
© E.B. Gordon, V.I. Matyushenko, V.D. Sizov, and B.M. Smirnov, 2008
in approaching the behavior of electron drifted in con-
densed media to that inherent to a gas [2,9].
Nevertheless it has been proved experimentally that
even at moderate (around kVs per cm) field the drifted
electrons can be accelerated up to energy of several eV
being sufficient for the matrix electronic excitation
[10,4,5]. However this fact itself does not facilitate the
solution for the problem of electrical discharge creation.
As the matter of fact the cross-section of electronic exci-
tation as well as a cross-section of any nonelastic process
has to be as high in a solid as in a gas. Thus at high densi-
ties typical for condensed matter the probability for drift-
ing electron to excite a matrix just after the threshold
overcoming is so large that at really accessible electric
field the electron additional overheating up to the ioniza-
tion threshold, i.e. 3 eV higher than the excitation thresh-
old, is absolutely negligible. This fact is independent of
the matrix temperature because the typical electron en-
ergy exceeds its thermal energy by two orders of magni-
tude. The probability of multi-step ionization is very low
as well because the electronic excitation of a matrix im-
mediately results in two-center trapped molecular-like
exciton [11] formation which in turn decays emitting VUV
photon, for short-living component the radiation time is as
short as 30 ns even for Xe [12]. The feasibility of trapped
exciton formation by electron impact in solid xenon has
been experimentally proved previously [13,14].
The absence of ionization in the course of electron
drift in condensed rare gases means the absence of elec-
tron multiplication being necessary for a discharge
self-maintaining. Moreover, positive ions are practically
motionless in a solid rare gas, so any electrical current
through a sample should quickly dump due to the com-
pensation of electric field by the accumulative space
charge. Such a problem may be bypassed by using the
electrical cell with photosensitive cathode [4]. Each elec-
tron creates many excitons during its drift through solid
rare gas, every of them decaying with VUV photon emit-
ting. This photon is able to cause the photoelectron effect
striking a cathode; these additional electrons entering the
matrix create the positive feedback that itself without ion-
ization could provide the multiplication of number of
electrons passed the sample and consequently the
self-sustained current through it [5].
In practice, the following approach [4,5] is more con-
venient. In a three-electrode cell the xenon crystal is
grown at the bottom, between the grid and anode. The gap
between cathode and grid is filled by inert gas — xenon,
argon or helium — at small pressure. In case of xenon the
sample temperature has been kept equal to 115–132 K to
have saturated vapor pressure in cathode-grid gap in the
range 7–40 kPa; while using more light gases they put into
cell at close pressure, the sample temperature was 77 K in
the case. In electric field applied to the cathode-grid gap
the electrons emitted from cathode enhance their energy
and then ionize gas forming the electron avalanche pro-
pagated from cathode to the grid with a speed above
107ñm/s. The electron number multiplication factor as
high as 103 could be achieved in a regime of exponential
growth, the Townsend discharge occurs at higher voltage.
Passing a grid and then entering the xenon crystal the
electrons drift there in voltage applied to grid-anode gap.
As it has been shown in experiment the drifting electron
passed the voltage of 1 kV induces emission of 20 VUV
photons (172 nm) [5]. The geometrical probability for
this photon to bombard the photosensitive cathode could
be estimate as 20% whereas the photoelectron emission
yield for zinc at � = 172 nm is 5·10–4 [15]. These esti-
mates demonstrate the possibility to achieve the condi-
tions of chain-branched electron multiplication being the
analog of self-sustaining electrical discharge regime. Un-
like common discharge the charge carrier multiplication
takes place in a gas whereas the positive feedback, being
necessary for self-sustaining operation, occurs in a solid
at the account of molecular-like exciton electrolumines-
cence caused photoelectrical effect on the cathode.
In principle self-maintaining electrical current in xe-
non could be realized as it was proposed in Ref. 2 without
any electron avalanche in a gas. However when metallic
cathode contacts directly with solid xenon the effect of
so-called back diffusion appears diminishing by orders of
magnitude the probability for photoelectron to enter the
crystal, CsI cathode is lucky exclusion because the proba-
bility of electron capturing by liquid xenon has shown to
be as high as 0.1 there [16].
In experiment with three-electrode cell we really ob-
served the transition to self-sustaining discharge (genera-
tion) that is characterized by appearance and subsequent
growth of regenerative peak in electron current, with its
maximum realized, as it follows from general consider-
ations, at the moment of positive feedback termination
[4,5]; the primary electrons were created by exposing the
zinc cathode to short pulse of 4th harmonics of Nd:YAG la-
ser irradiation (266 nm). The positive feedback duration
was naturally equal to the time of excess electron drift
through the xenon crystal. That time was known with good
accuracy because in the field range 102–105 V/cm the elec-
tron drift velocity in condensed xenon did not depends on
the field strength and was equal to 2�105 cm/s [8].
Regretfully, as in any experiment where electrons
move through solid dielectrics, there is a limit for the total
charge flowing through the sample that stops the current
through the sample. That is due to molecular admixtures
that capture electrons and create a motionless volume
charge. Any purification method, including the effective
electro-spark technique [17] we used, can not diminish
the electronegative impurity (oxygen first of all) content
below 0.2 ppb, that corresponds to molecular density in a
1204 Fizika Nizkikh Temperatur, 2008, v. 34, No. 11
E.B. Gordon, V.I. Matyushenko, V.D. Sizov, and B.M. Smirnov
solid of 6·1012 ñì–3. All these molecular traps capture the
electrons provided the current exists enough long. In our ge-
ometry the corresponding charge density is sufficient to
compensate the fields with the strength exceeding
100 kV/cm. That is why a scientist used to work with as low
current as possible within the time when the current is still
not terminated. Then, in order to continue the experiment
one should remelt the sample. In our conditions the limiting
time of «discharge» existence was only tens of seconds.
It is turned out however that provided all resistors re-
stricting the current in electrical circuit were removed the
compensation of electric field by accumulating space
charge did not exist anymore and the discharge initially
triggered by laser pulse may exist for ever. Such a behav-
ior we explained hypothetically [4] by trapped electron
blowout under impact with fast drifting electron. The
electron affinity to molecular oxygen is only 0.43 eV, so
the cross-section of the electron detrapping by impact of
free electron having sufficient energy should be enough
high [18]. Thus starting from some density of excess elec-
trons the steady concentration of motionless negative
charges (defined by equilibrium between free electron
trapping to impurity and negative ion decomposition by
the impact with free electron) may become to be less than
that compensating the voltage which induces a drift.
This work is devoted to testing that hypothesis, to real-
ization of steady bulk «discharge» in solid xenon, as well
as to studying this phenomenon being peculiar and attrac-
tive for the applications, especially for revealing the xe-
non matrix transformation at very high pressure. The
problem of the discharge initiation is to be cardinal be-
cause due to quadratic dependence of the detrapping rate
on free electron density a steady discharge can not be
achieved spontaneously.
Though the principal ability for excess electron to drift
in nonpolar solid as it does it in a gas is obvious, the feasi-
bility for an electron to acquire the energy sufficient for
the new charge producing in reasonable electric field is
the specificity of a condensed matter. In dense gas such a
process will be simply impossible (as the experiments
have shown [19] the discharge threshold starting from gas
pressure of 100 bar roses to the electric field strength
practically unattainable due to breakdown occurence).
Quite natural and important from viewpoint of possible
application at high and ultrahigh pressure is the question
what should happen at atomic densities higher than those
characteristic for solid and liquid states at normal condi-
tions. In other words, one should distinguish whether
high mobility of excess electron is a consequence of acci-
dental coincidence of the density of condensed heavy rare
gases at normal conditions — about 3�1022 ñm–3 — with
the density when the compensation of an electron attrac-
tion and repulsion to atomic system provides the weak-
ness of its interaction with media [20], or that high mobil-
ity has to last in wide range of densities, and consequently
pressures. Of course, high pressures should induce the
other effects as well and they will be analyzed in this
work too.
3. Experimental setup
The design of the technique built up in IPCP RAS is
shown in Fig. 1; it is close to that described in Ref. 4.
Only the materials and tools destined for ultrahigh vacu-
um operation have been used in assembly. All tubes were
from stainless steel, the Varian flanges and copper gas-
kets, as well as Swagelok ferrules, pipes and nuts were
used for connectors. In construction of both discharge
cell and xenon purification chamber any plastics, glues,
and even silver soldering were avoided, last due to impos-
sibility to remove the traces of flax; only indium (for seal-
ing), sapphire, glass, Kovar, and Macor ceramics were
applied as materials. Whole vacuum system could be de-
gased by heating up to 200�C. The Varian Turbo-V70
turbomolecular pump supported by Varian SH-100 scroll
pump were used, the stainless steel gas receivers had spe-
cial fingers for xenon condensation by liquid nitrogen.
Pressure was measured by the modification of Barocell
pressure gauge compatible with high vacuum.
Necessary efficiency of xenon purification from
electronegative impurities — about 10–10 — can not be
kept for a long time in separate cylinder. Thus the purifi-
cation system periodically retrieved the necessary purity
was an incorporated part of setup. The electrospark
method of xenon purification has been used as previously
[4]. The titanium blade-shaped ring served as cathode and
steel cylinder was used as anode, interelectrode distance
was 0.2 mm. Before operation the purification chamber
has been submerged to the liquid ethyl alcohol perma-
Low-temperature electrical discharge through solid xenon
Fizika Nizkikh Temperatur, 2008, v. 34, No. 11 1205
1
2 3
4
5 5
6
7
9 11
10
12
885585
7
Fig. 1. Setup scheme: turbomolecular pump (1); scroll pump (2);
vacuum valve (3); Pirani pressure gauge (4); ultrahigh vacuum
valves (5); Barocell pressure gauge (6); stainless steel cylin-
ders with finger (7); high vacuum valves (8); purification
chamber (9); stainless steel cylinder (10); experimental cell lo-
cated inside of cryostat (11).
nently cooled down to porridge-like consistence by add-
ing liquid nitrogen. Liquid xenon interface was 1 cm
higher than cathode level. The homemade high voltage
power supply had manipulated output from 1 to 10 kV.
Because the frequency of 75 kHz has been rectified there
the output capacitor intended for smoothing the signal
had a low capacity and as a result the output energy was
insufficient for the spark development — at high
anode-cathode voltage an electric arc simply appeared in-
side liquid Xe. With the additional 0.1 mF outlet capacitor
one or two sparks exist, moving around the interelectrode
gap as metal sputtered, starting from some voltage ap-
plied — at 1.5 kV for nonpurified xenon and at 8 kV at the
end of procedure. The spark current has been limited by
powerful 1 M� resistor. The cavitation accompanied the
electrical breakdown provided intensive intermixing of
liquid xenon supplying the spark region with new portion
of a liquid. Several hours of such a procedure were suffi-
cient to diminish the contamination of Xe down to 0.2 ppb
in a few cc of liquid.
Liquid nitrogen optical cryostat has been designed es-
pecially for this study in the Institute of Solid State Phys-
ics RAS. It possesses 4 arrays of sapphire optical win-
dows suitable as for UV range where excitonic emission
takes place as in visible and near IR ranges. The complex-
ity of perfect xenon crystal growing consists of small, less
than 2 K, gap between the temperatures of Xe melting
and boiling. Due to large volume of the sample and conse-
quently large heat releases under gas xenon condensation
and then its solidification, the common technique of
temperature stabilization with feedback was inapplicable.
The system of stabilization we chosen acts as follow. The
homemade controller kept the excess pressure of vapor in
liquid nitrogen jacket constant and equal to 0.2 bar. The
main bath of the cryostat where the experimental cell was
situated has been connected to ambient air through a nee-
dle-valve controlled the rate of gas outflow and such a way
the supply of liquid nitrogen from the jacket to main bath
through a capillary, the necessary flow has been deter-
mined experimentally. The temperature of the cell was
measured by copper-constantan thermocouple. At the mo-
ment when necessary amount of liquid xenon has been
condensed to the cell it was sealed from the supply system
by the valve and the outflow controlled by needle valve
was strongly diminished (to compensate the sharp cessa-
tion of heat release at the end of condensation). When in a
few minutes the temperature reached the Xe triple point
T = 161 K the needle-valve has been closed at all, the crys-
tallization proceeded isothermally and frontally from the
down to the top during about 20 min. After the crystalliza-
tion completion the cooling rate was quickly enhanced and
after the temperature becomes to be 77 K whole cryostat
bath was filled by liquid nitrogen. The quality of the crys-
tals such grown was better then previously [4].
In this work the new design of three-electrode experi-
mental cell has been used, the photo is presented in Fig. 2.
Its body was the sapphire tube 1 with 28 mm inner diame-
ter gripped through indium sealing between two steel
flanges 2 and 3, the possibility of optical observation both in
UV, visible and IR ranges has been achieved in such a way.
To compensate mechanical tension appeared under tempera-
ture changes and being large enough due to the diversity of
materials applied in construction, the flanges were pressed
to sapphire tube through flat spring 4 as it has been done in
the cell meant for work with solid helium [21].
The bottom flange 2 served as anode and its working sur-
face 6 was uplifted on 3 mm to avoid the electrical break-
down from indium spew, sqeezed out inside the cell under
sealing, to grid 8 along the side surface of a crystal. Two
40 �m tungsten wires were installed at the center of anode;
they rose on 0.5 mm above its surface. Their task was to give
rise powerful spark initiated the main discharge.
The zinc cathode 10 mm in diameter was attached to
upper flange. The distance between cathode and grid
made of stainless steel mesh with 60% geometrical trans-
parency was 2 mm and their electrical insulation has been
attained by figured gasket made of Macor ceramics. By
changing the length of cathode holders the distance
1206 Fizika Nizkikh Temperatur, 2008, v. 34, No. 11
E.B. Gordon, V.I. Matyushenko, V.D. Sizov, and B.M. Smirnov
9
4 3
1
5
7
8
6
2
Fig. 2. Experimental cell: sapphire tube (1); bottom flange (2);
upper flange (3); stainless steal flat spring for thermal expansion
compensation (4); Caprolan insulator (5); anode surface (6); mesh
grid (7); the position of zinc cathode inside of Macor gasket (8);
high-voltage feedthrough (9).
between grid and anode determined the maximal sample
thickness could be changed from 2 to 4 mm. The cell was
provided with electrical heater rolled around upper flange
and with thermocouple.
Unless a discharge was terminated by space charge
captured by the impurities presented inside a sample, its
power was so large that it could melt whole crystal during
few seconds. For this reason the 430 pF capacitor was
used as voltage supply. It was charged by the power unit
through 110 k� resistor; the discharge current has been
restricted by 1 k� resistance. The same scheme was used
for grid-anode feeding, 0.1 �F capacitor was charged
through 10 M� resistor in the case. Owing to this the dis-
charge represents the sequence of short pulses; the repeti-
tion rate dependent on power unit voltage was usually
chosen about 1 Hz.
4. Experimental results
The different regimes of electric discharge through
solid xenon are illustrated by Fig. 3. The visible glowing
was extremely intensive — all photos were made at day-
light and background looks like black in the presence of
discharge only by contrast. The shape of xenon sample is
shown in upper photo. Only spark breakdown takes place
at small voltage U applied to gas gap, as it is seen from the
next photo. The observation through the mutually perpen-
dicular windows manifests the spark to be just at axes of
the cell and crystal. The temporal profile of the current in
cathode-grid gap at such a condition is shown in Fig. 4,
and the «spark» duration equal to 300 ns together with its
high brightness and comparably large for a spark radius
indicate that is not simple breakdown; namely the notice-
able part of its trace an electron does not moved in a crack
but drifts through a crystal bulk nearby.
Beginning from some threshold voltage U, extremely
bright uniform glow sharply appears in whole sample vol-
ume, the pulse repetition rate does not change at this mo-
ment. The pattern of this discharge shown at Fig. 3,c is the
same being observed through the windows from all four
sides of the cell, no extra intensity was found both at the
center there the spark has been before and at the edges of
crystal, last is the evidence for the absence of specific role of
sample surface in emission. The durations of both light and
current pulses were almost two orders of magnitude more
long than those in the regime of spark breakdown (look at
correspondent oscillogram in Fig. 4). The charge re-estab-
lishment time for the feeding capacitor in cathode-grid cir-
cuit is rather short �c = ÑR = 0.43�10–9 F�1.1�104 � = 50 �s;
thus the observed discharge repetition rate (about 1 Hz) is
definitely controlled by the charge of capacitor in the
grid-anode circuit, where re-establishment time really is
�c = ÑR = 1�10–7F�1�107 � = 1 s. According to voltmeter
readings the capacitor is loosing practically all its charge to
the end of every discharge pulse; it means that the charge
flowed through the grid-anode gap filled up with solid xe-
non q = CU = 1�10–7F�2.4�103 V = 2.4�10–4 C is much
more than that flowing through the cathode-grid gap filled
up with a gas q = CU = 0.43�10–9F�1�102 V = 4.3�10–6 C.
In other words, the additional charge carriers — electrons
— in amount being 50 times more than that in a gas gap
are formed in the grid-anode gap. Since only electrons es-
Low-temperature electrical discharge through solid xenon
Fizika Nizkikh Temperatur, 2008, v. 34, No. 11 1207
a
b
c
d
Fig. 3. The photos of different regimes of electron current
through solid xenon at constant grid-anode voltage, Uga = 2.4 kV,
and different cathode-grid voltages, Ucg: a) Ucg = 0 (no cur-
rent); b) Ucg < 100 V (only spark); c) 100 V < Ucg < 200 V
(uniform discharge); d) and Ucg > 200 V (weak glow).
Tek Run Sample
Tek Preview Sample
R1
R1
a
b
Ref1 2.0V 4.0us
Ref1 2.0V 4.0us
M 200ns 2.5GS/s 400ps/pt
A Ch1 0.0 V
M 4.0us 125MS/s 8.0ns/pt
A Glitch Ch1
Fig. 4. Oscillograms of visible emission intensities in a)
«spark» regime, time scale 0.2 �s/div, and b) discharge regime,
time scale 4 �s/div.
caping a cathode are then multiplied in a gas the effect
hardly could be attributed to less efficient photoelectric
effect on the grid surface. Most likely the reason is the ad-
ditional electron multiplication in the narrow gap be-
tween the grid and upper surface of crystal: taking into ac-
count the high xenon dielectric constant equal to 1.8 the
electric field there is 40 times more than in cathode-grid gap.
For achieving the additional multiplication factor in electron
avalanche equal to 50 it is sufficient to have a gas slit between
a grid and upper crystal face only 0.10–0.15 mm.
Rather low electrical resistance of a crystal follows
from discharge pulse parameters observed — since the
discharge duration of 4 �s is sufficient for 0.1 �F capaci-
tor depletion, the impedance should be less than 40 �. In
accordance with that estimate the introduction of 1.3 k�
resistor into discharge circuit entirely kills the discharge
through solid xenon. Meantime the gas gap impedance
was not less then 15 k�. The analysis of light emission
pulse presented in Fig. 4 leads to the interesting conclu-
sion: there are few successive electrical breakdowns of
solid sample. The moments of sparks appearance are seen
due to high-frequency noise they induce at the back-
ground of a signal detected by photodiode. Every spark
gives birth to its own electron avalanche in a gas gap and
then to its own «discharge» through solid xenon. Every
discharge such induced represents the pulse having the
maximum at the electron time-of-drift through the sample
being equal in our case to 2 �s. Such a behavior is typical
for any generator: the additional pulsed excitation of a
generator already working in a steady regime causes a
perturbation existed during the time of positive feedback
action. The same shape had the regenerative peak of dis-
charge current we have observed in vicinity of the thresh-
old of self-sustained discharge through solid xenon [4,5].
Such effects are quite naturally more significant near the
vertical axes of a sample where initial breakdown takes
place. Thus they are less indicative in current signal, that,
contrary to emission collected by lens from the central
part of a sample, represent total electron current through a
sample.
The most convincible argument in favor of such inter-
pretation of the experimental data is the discharge behavior
under further growth of voltage applied to cathode-grid
gas gap. Both current and light emission are practically
invariable while voltage grows from 100 to 200 V, but
then the crystal glowing intensity falls down step-wisely
and becomes comparable with emission of gas discharge
in cathode-grid gap (see Fig. 3,d). Due to the small dis-
charge current at that stage it can be feed continuously
through the 10 M� resistor — the glowing observed be-
comes continuous-wave indeed. Such a behavior saves at
further voltage growth above 200 V. The threshold volt-
age values presented above are dependent of course on
the sample thickness.
The unusual effect of discharge depression whereas
the voltage applied to gas gap and consequently the factor
of electron multiplication in avalanche both grow can be
rationalized as follows. Beginning from cathode-grid gap
voltage 200 V the continuous-wave discharge through gas
gap appeared — it was observed at close voltage in ab-
sence of grid-anode voltage as well. It causes effective
permanent bombardment of a crystal by electron escaping
gas discharge area. Such pre-ionization of a crystal be-
comes to be high enough for suppressing the spark devel-
opment. As a result, only weak discharge, controlled by
the space charge of impurity negative ions, took place.
Beside VUV luminescence from excitons the intensive
emission in UV and visible has been registered from uni-
form discharge through solid xenon. The overview spec-
trum is presented in Fig. 5. The main lines there belong to
xenon, they are slightly broadened and (about 0.4 nm)
shifted to the blue from their gas phase positions. It is
very interesting that we observed all lines of ions Xe+
usually present in gas xenon discharge but no line of Xe
atom has been registered including these possessed high
intensity in both high and low pressure xenon gas dis-
charges [22,23]. As we already mentioned, the direct ion-
ization of a matrix is impossible in our conditions, just
preliminarily excited species could be the precursors of
1208 Fizika Nizkikh Temperatur, 2008, v. 34, No. 11
E.B. Gordon, V.I. Matyushenko, V.D. Sizov, and B.M. Smirnov
220 240 260 280 300
0
4
8
12
16
In
te
n
si
ty
,
ar
b
.
u
n
it
s
350 400 450 500 550
0
1
2
3
In
te
n
si
ty
,
ar
b
.
u
n
it
s
a
b
Wavelength, nm
Wavelength, nm
Fig. 5. Overall spectra of solid xenon in visible (a) and UV (b)
regions. No correction for spectral sensitivity of photomulti-
pliers has been made. Xe crystal was kept in equilibrium with its
vapor at 132 K, thus the cathode-grid gap was filled by xenon
with pressure around 4.5 kPa. The width of individual lines regis-
tered at low monochromator scanning rate was 0.4 nm.
ions. Since only excited species are molecular-like
excitons the process
Xe Xe Xe2
�
�
should be rather effective under hot electron impact or
VUV quantum absorption. Note that unlike gas phase
where negative ions Xe
do not exist, the excess electron
potential energy is negative in solid xenon lattice.
Another peculiarity of solid xenon is the narrowness
of the exciton emission band: while in a gas and in a liquid
the intensive luminescence arises already at 200 nm we
see no traces of emission at 190 nm.
Extra spectral lines and continuums presented in spec-
tra of discharge in solid xenon may belong to impurities.
Though they regretfully related to unknown centers, the
high intensity of emission from trace species with relative
content about 10–10 is good evidence for high probability
of electroluminescence connected with the fact that real
path of drifting electron in a sample is 103 times longer
than the sample thickness [4].
That is especially interesting for spectroscopic and
other physical and chemical applications that the method
of organization of electric discharge through solid xenon
is the brute force technique: no external factors are neces-
sary for realization. One should only prepare the sample
and apply the electric field of sufficient strength to it. If so
the discharge can be developed without any optical access
to a sample, in particular in the experiments with extra
high pressure and in blasting operations [24].
Such a way the universal approach has been developed
allowing, on the one hand, introducing excess electrons to
solid xenon and, on the other hand, studying the interactions
of hot electrons with practically any molecules isolated in
inert matrix. The temperature can be arbitrary low provided
the helium will be used as the gas in cathode-grid gap.
5. About the discharge through solid Xe at high and
ultrahigh pressures
The method under consideration may be applied in
principle to high-pressure studies of the behavior of ex-
cess electrons in solid xenon. As it was shown in Ref. 25,
the current of excess electrons proceeds in a space along
the tubes whose axes are Voronoi–Delone channels. This
is additional to the classical description of the electron
behavior in metals on the basis of the Fermi surfaces and
exhibits that the metal conductivity is not a result of the
crystal structure of metals only. That is of obvious interest
to investigate the excess electron current at pressure be-
ing high enough to modify electron potential inside a
solid. However, even the very feasibility to organize a
discharge at high pressure is not quite clear and both
theoretical and technical analysis is required.
Thus the simplest approach, applicable however only
for blasting experiments, is the exposition to shock-wave
compression of xenon crystal with preliminarily sustained
electrical discharge. Because the duration of a discharge in
a sample of 1 cm thickness is about 5 �s one has a time for
pressing. The presence in a sample of excess electrons
makes it possible to study the electron mobility far below
the pressures provided effective ionization of a matrix.
The realization of the method in the condition of static
compression looks much more problematic. First of all in
the design of an experimental cell the gas gap above the
crystal should be removed because the discharge through
high pressure gas is impossible. However the experience of
present experimental work shows that this difficulty may
be bypassed by using two-electrode cell with CsI cathode.
The effect of back diffusion is absent here, and the absence
of electron multiplication in gas avalanche will be compen-
sated by higher photocathode efficiency and removing the
screening of electroluminescence by mesh grid.
In principle beginning from high enough pressure the
chemical potential of excess electrons in solid rare gas be-
comes positive, so thermal electron could not enter the
sample then. However our estimates show that it is impor-
tant only for more light rare gases, for xenon such a pres-
sure is about 130–150 GPa. Such a way this effect does
not prevent a static operation, in particular in a frame of
technique based on diamond anvils [26].
For the studies at static pressure the crucial question is
the possibility for an excess electron to be accelerated in a
solid at high pressure, the exciton and molecular ion
structures and energies are of great importance in this
case.
We now examine briefly the properties of compressed
xenon which should be analyzed in future work. At very
high pressures and consequently at very small distances
between nearest neighbors, the electron band of valence
electrons intersects the boundary of continuous spectrum,
and then xenon becomes a metal. According to experi-
mental study [27,28], the dielectric-metal transition for
xenon occurs at the pressure of 130–140 GPa. But consid-
erably below this pressure the gap between the electron
valence band and the boundary of continuous spectrum
decreases to form a broad dielectric–semiconductor–me-
tal transition.
The study of intermediate semiconductor state of
xenon would give a more profound understanding the
dielectric–metal transition. But in practice the conducti-
vity in the semiconductor regime should be entirely de-
fined by electronegative impurities and not by own matrix
ionization. Only a deep (<10–7%) xenon purification
together with injection excess electrons into the matter
would remove this trouble.
In principle, two types of the electron excitation,
atomic and molecular, are possible in condensed inert
Low-temperature electrical discharge through solid xenon
Fizika Nizkikh Temperatur, 2008, v. 34, No. 11 1209
gases at moderate pressures. In reality, only molecular
exciton emission is observable in solid xenon with a spec-
tral band position close to that of Xe 2
� excimer band in a
gas. Meantime, the molecular spectral band should disap-
pear at enough high pressure because small distances be-
tween the excited atom nearest neighbors exclude exis-
tence of its bound state with one atom. The corresponding
pressure and the character of exciton band distortion and
decay are of obvious interest.
We now enumerate the effects appeared in solid xenon
as the pressure increases and revealed themselves as the
changes in excess electrons mobility and excitonic lumi-
nescence characteristics.
First of these is a drop of the mobility of excess electrons
when a pressure increases above atmospheric one at which
the reduced mobility of excess electrons in solid xenon ex-
ceeds by three-four orders of magnitude that in gases of the
same pressure [19]. That high electron mobility can be ex-
plained by existence of Voronoi–Delone channels between
xenon atoms, where a slow electron propagate [20]; these
regions will disappear at high pressures resulting in signifi-
cant decrease of slow electrons mobility.
The next effect consists in strong distortion of exci-
tonic luminescence band shape. At atmospheric pressure
molecular-like exciton emission is close to that of Xe 2
�
excimer molecule in a gas due to large difference between
interatomic distance in two-center exciton and that in a
matrix. But a pressure increase leads to a decrease in the
distance between nearest neighbors of solid xenon, and at
certain pressure the molecular-like spectrum will be re-
placed by the atomic-like one. All these effects are ex-
pected already in the kbar pressure range.
At higher pressure the effect of diminishing the energy
necessary for atomic ionization in a matrix becomes im-
portant. That is the consequence of two effects. On the
one hand, the system electronic ground state goes up due
to exchange interaction between valence electrons, and
on the other hand, the level of ionic state goes down. But
last effect is of another nature than that for the ground
state. Indeed, the electron term of the ionized state in av-
erage does not change but splits into many levels trans-
formed to expanded band. Both effects lower the ioniza-
tion energy and so-called «ionization by pressure»
proceeds of course via lower edge of ionization band. Of
course, excitonic emission should disappear at that pres-
sure as a result, and then the ionization of the matrix by
excess electrons accelerated in electric field may proceed.
This effect can display the threshold behavior and in this
case it could be interpreted as insulator–semiconductor
phase transition.
Beginning from such a pressure together with the cur-
rent of excess electrons the conductivity of pure matrix
should take place, especially in blasting experiments
where the temperature of compressed matter is high
enough and large amounts of intrinsic free electrons
appear at pressures significantly lower than those corre-
sponded to metallization, i.e. forbidden band disappear-
ance. The mobility of the electrons may be at the same
time still rather low, and only at high inner ionicity the
character of bonds should vary from van der Waals bind-
ing to metallic one. New ordering such appeared will re-
store high mobility up to that inherent to a metal.
Probably in the same range of pressure the electron
will loose its ability to penetrate inside solid xenon from
gas phase because the electric potential inside solid xenon
for a slow electron becomes higher than that outside it.
6. Conclusion
The experimental approach just elaborated allows the
creation of stable uniform electrical discharge inside
solid xenon. High energy electrons make it possible to ex-
cite the xenon matrix as well as the impurities stabilized
inside solid xenon and to cause their chemical reactions in
inert matrix at arbitrary low temperature. The method
luckily can be realized at high and ultrahigh pressure as
well. The feasibility of introducing large amounts of ex-
cess electron into compressed solid xenon assures the ex-
perimental observation of several pronounced effects.
Their characteristics will be reliable base for detailed the-
oretical analysis and, that is especially important, for un-
derstanding the matter transformation at high pressure at
all. The design of the technique allowing the excess elec-
tron introducing into solid xenon being compatible with
conditions of blasting experiment is now in progress.
This work is supported partially by Russian Founda-
tion for Basic Researches grant 07-03-00393.
1. J.S. Townsend, Electrons in Gases, London, Hutchin-
son’ssci. Techn. Publ. (1947).
2. E.B. Gordon, V.V. Khmelenko, and O.S. Rzhevsky, Quan-
tum Electronics 21, 227 (1994).
3. E.B. Gordon, V.V. Khmelenko, and O.S. Rzhevsky, Chem.
Phys. Lett. 217, 605 (1994).
4. E.B. Gordon, G. Frossati, and A. Usenko, J. Exp. Theor.
Phys. 96, 846 (2003).
5. A. Usenko, G. Frossati, and E.B. Gordon, Phys. Rev. Lett.
90, 1532017 (2003).
6. E. Aprile, K.L. Giboni, and C. Rubbia, Nucl. Instrum.
Meth. Phys. Res. A24l, 62 (1985).
7. M. Ichige, E. Aprile, T. Doke, K. Hasuike, K. Itoh, J.
Kikuchi, and K. Masuda, Nucl. Instrum. Meth. Phys. Res.
A333, 355 (1993).
8. G. Ascarelli, J. Phys. Chem. 84, 1193 (1980).
9. B.M. Smirnov, Phys. Uspekhi 45, 1251 (2002).
10. A.S. Schussler, J. Burghoorn, P. Wyder, B.I. Lembrikov,
and R. Baptist, Appl. Phys. Lett. 77, 2786 (2000).
11. N. Schwentner, E.E. Koch, and J. Jortner, Electronic Exci-
tations in Condensed Rare Gases, Vol. 107 of Springer
Tracts in Modern Physics, Springer-Verlag, Berlin (1985).
1210 Fizika Nizkikh Temperatur, 2008, v. 34, No. 11
E.B. Gordon, V.I. Matyushenko, V.D. Sizov, and B.M. Smirnov
12. E.V. Savchenko and I.Ya. Fugol, in: Kryokristalli, B.I.
Verkin and A.F. Prikhod’ko (eds.) (in Russian), Kiev,
Naukova Dumka (1983).
13. I.Ya. Fugol, O.N. Grigorashchenko, and E.V. Savchenko,
Phys. Status Solidi B111, 397 (1982).
14. I. Reimand, E. Gminder, M. Kirm, V. Kisand, B. Steeg, D.
Varding, and G. Zimmer, Phys. Status Solidi B214, 81 (1999).
15. Handbook of Chemistry and Physics, Edition 86, D.R. Lide
(ed.), London: CRC Press (2003–2004).
16. A. Breskin, Nucl. Instrum. Meth. Phys. Res. A371, 116
(1996).
17. S.G. Pokachalov, M.A. Kirsanov, A.A. Kruglov, and I.M.
Obodovski, Nucl. Instrum. Meth. Phys. Res. A327, 159
(1993).
18. B.M. Smirnov, Negative Ions, McGraw-Hill, New York,
London (1982).
19. E.B. Gordon and A.F. Shestakov, Fiz. Nizk. Temp. 27,
1192 (2001) [Low Temp. Phys. 27, 883 (2001)].
20. E.B. Gordon and B.M. Smirnov, J. Exp. Theor. Phys. 98,
924 (2004).
21. E.B. Gordon, T. Kumada, M. Ishiguro, and Ya. Aratono, J.
Exp. Theor. Phys. 99, 776 (2004).
22. E.Kh. Baksht, M.I. Lomaev, D.V. Rybka, and V.F. Tara-
senko, Quantum Electronics 36, 576 (2006).
23. A.K. Malik, P. Montarde, and M. G. Haines, J. Phys. D:
Appl. Phys. 33, 2037(2000).
24. V.E. Fortov, V.Ya. Ternovoi, M.V. Zhernokletov, M.A.
Mochalov, A.L. Mikhailov, A.S. Filimonov, A.A. Pyalling,
V.B. Mintsev, V.K. Gryaznov, and I.L. Iosilevskii, J. Exp.
Theor. Phys. 97, 259 (2003).
25. E.B. Gordon and B.M. Smirnov, J. Exp. Theor. Phys. 101,
186 (2005).
26. P.W. Bridgeman, Proc. Am. Acad. Arts Sci. 72, 207 (1938).
27. K.A. Goettel, J.H. Eggert, and I.F. Silvera, Phys. Rev. Lett.
62, 665 (1989).
28. R. Reichlin, K.E. Brister, A.K. McMahan, M. Ross, S.
Martin, Yo.K. Vohra, and A.L. Ruoff, Phys. Rev. Lett. 62,
669 (1989).
Low-temperature electrical discharge through solid xenon
Fizika Nizkikh Temperatur, 2008, v. 34, No. 11 1211
|