Electrical discharge plasma characteristics in pure Ar gas
at multi-atmospheric pressure using the automatically
pre-ionized plasma electrode
July 2005
Department of Energy and Materials Science Graduate School of Science and Engineering
Saga University Sung-Ki Hong
© 2005 by Sung-Ki Hong
All right reserved.
- I -
ACKNOWLEDGEMENTS
I am thankful to my advisor, Professor Chobei Yamabe for his guidance, support
and cooperation throughout the course of my education. This work is the result of his
constant inspiration, encouragement and words of wisdom.
My sincere thanks to the members of my dissertation committee, Professor
Hiroharu Fujita, Professor Saburoh Satoh and Professor Kazuhiro Muramatsu for
their thoughtful comments.
I am grateful to the Ministry of Education, Science, Sports and Culture, Japan for
their financial support. Further, I am thankful to Dr. Satoshi Ihara and Dr. Nobuya
Hayahsi. Special thanks to Professor Sang-Bong Wee and Dr. Hee-Sung Ahn.
I am also very grateful to my family for their invaluable affections and
encouragement during our stay in Kyushu, Japan.
Finally, I thank my wife Yong-Jin Chun for her patience, devotion and
understanding for the past 3 years.
Saga University, Saga, Japan
June 20, 2005
Sung-Ki Hong
홍 성 기
- II -
ABSTRACT
Electrical discharge plasma characteristics in pure Ar gas at multi-atmospheric pressure using the automatically pre-ionized plasma electrode
Sung-Ki Hong
B.S., Korea University of Technology and Education
M.S., Korea University of Technology and Education
Ph.D., Saga University
In this study, for the investigation of the possibility of the Ar2* excimer laser
action and the development of the electrode structure for the efficient transversely
excited (TE) gas laser, a new pre-ionization electrode was designed using the
surface-corona pre-ionization method. It was investigated experimentally in terms of
generated charge density, electrical characteristics. These experimental results
suggest that a sharp edge of the ground electrode is possible to radiate the strongest
UV light. Therefore, the Automatically Pre-Ionized (API) plasma electrode system
was designed using a new pre-ionization electrode by the surface-corona pre-
ionization method. Characteristics of the main-discharge in multi-atmospheric (1~5
atm) pure Ar gas were investigated using the automatically pre-ionized plasma
electrode. The uniform main-discharge was formed and its volume and the
breakdown voltage increased with increasing Ar gas pressure. The instantaneous
maximum discharge electric power was 90 MW at 5 atm Ar gas and the maximum
energy deposition was 1.4 MW/cm3. It was demonstrated that the uniform pre-
ionization formed the uniform main-discharge by the control effect of Cpr and the
maximum energy deposition was increased.
- III -
The control effect of Cpr was examined the time dependent main-discharge from
two-dimensional simulation of electric field distribution of the automatically pre-
ionized plasma electrode discharge system. In addition, the light emission
characteristics of the discharge system using the automatically pre-ionized plasma
electrode were investigated by the measurement of Ar atomic line and Ar ionic line.
The intensity of the line at 427.8 nm increases proportionally to the approximate
cubic of the Ar gas pressure (Y = 0.22X3, where Y is intensity and X is the Ar gas
pressure). On the other hand, the line intensity at 696.9 nm shows saturation at
pressure above 4 atm. These experimental results suggest that the lasering band
Ar2*(1Σu) is enhanced due to the increase of the excited Ar* atom which is formed
by the electron collision reaction Ar + e→Ar* + e and the Ar2*(3Σu) is saturated by
the recombination process. On the basis of such experimental results, the new
resonator for the Ar2* excimer laser action was designed.
- IV -
TABLE OF CONTENTS
page ACKNOWLEDGEMENTS ....................................................................................... I ABSTRACT................................................................................................................II 1. Introduction............................................................................................................ 1
1.1 The electron beam pumped method ................................................................ 3 1.2 The discharge pumped method ....................................................................... 4 1.3 Aim of this study ............................................................................................. 8 Reference............................................................................................................... 9
2. Theoretical considerations ...................................................................................11
2.1 Rare gas kinetics considerations ................................................................... 12 2.2 Discharge formation and stability in high pressure gas ................................ 15 2.3 General characteristics of the glow (or uniform) discharge.......................... 22 2.4 Excitation circuits.......................................................................................... 26 2.5 The main discharge ....................................................................................... 28 Reference............................................................................................................. 31
3. The pre-ionization source for excimer laser ...................................................... 32
3.1 Experimental set-up and methods ................................................................. 33 3.2 Experimental results and discussions............................................................ 35 Reference............................................................................................................. 38
4. The Automatically Pre-Ionized plasma electrode discharge system ............... 39
4.1 Experimental set-up ...................................................................................... 39 4.2 Circuit characteristics.................................................................................... 43 4.3 Experimental results and discussions............................................................ 45
4.3.1 Discharge characteristic of the API plasma electrode ......................... 45 4.3.2 Discharge characteristic of the API discharge system without Cpr...... 48 4.3.3 Using the plate electrode as an anode.................................................. 52
- V -
Reference............................................................................................................. 54 5. The electric field of API plasma electrode system............................................. 55
Reference............................................................................................................. 66 6. Characteristics of discharge pumped Argon gas excitation............................. 67
6.1 Light emission characteristics of discharge pumped Ar gas ......................... 70 Reference............................................................................................................. 78
7. Conclusions........................................................................................................... 80 Appendix #A Necessary conditions for the discharge-pumped Ar2* laser ......... 82 Appendix #B Design of a new chamber and electrode ......................................... 84 List of publications................................................................................................... 92
Ch.1 Introduction 1
1. Introduction
There have been considerable demands for the development of compact short
wavelength lasers in the vacuum ultraviolet (VUV) spectral region. Such compact
short wavelength lasers would be applicable to various scientific and industrial fields,
such as photochemistry, biological science, and new types of materials processing.
Recently more attention is paid to short wavelength lasers in the VUV as coherent
light sources in the future optical lithography industry. Optical lithography is
considered the most desirable technique for mass fabrication of advanced
semiconductor devices. Currently available practical compact VUV lasers are the
ArF excimer laser at 193 nm and the F2 laser at 157 nm, both of which are excited by
a compact discharge device. Possible path of lithography technologies is shown in
Fig. 1.1.
KrF(248nm) ArF(193nm)
SCALPEL(Scattering with Angular Limitation
Projection Electron beam Lithography)
(100keV electron)
Ar2*(126nm)
EUV(Extreme UltraViolet Lithography)
(13nm)
XRL(X-ray Lithography) (1nm, 1x proximity)
F2(157nm)
IPL(Ion Projection Lithography)
(75keV He+ ion)
NGL(Next-Generation Lithography)
<50nm
~110 or 90nm ~90 or 70nm ~50nm
Fig. 1.1 Possible path of lithography technologies
Ch.1 Introduction 2
∑∑ ++− g
1u
1,3
In optical lithography, resolution is given by the equation
NAkR λ
1esolution = (1.1)
where λ and NA are the exposure wavelength and numerical aperturea of the optical
lithography tool, and k1 is a constant for a specific lithographic process. As the
wavelength becomes shorter, the light source becomes more complex and expensive.
Turning now to optical devices, rare-gasb excimer lasers are a vacuum ultra-
violet (VUV) laser, achieved laser action from the transitions of the excited
dimer Ar2*(around 126 nm), Kr2*(around 146 nm), and Xe2*(around 172 nm). The
emission wavelength of Ar2* is 126 nm which is long enough to use transmission
optical elements such as MgF and LiF. The Kr2* laser has an even longer emission
wavelength centered at 147 nm which relaxes the conditions for optics and would
become a competitor to the F2 laser at 157 nm. From the practical viewpoint, rare
gases are chemically inert which shows a contrast to chemically active fluorine used
in the F2 laser. Among these rare gas excimers, Ar2* excimer laser produces radiation
with the highest photon energy of 9.8 eV. An electron beam excitation has been the
only excitation method to oscillate the Ar2* excimer laser [1-4]. However, the
electron beam excitation method requires a rather large facility to be operated,
resulting in a low average power of the VUV laser with a low repetition rate. The
electron beam excitation method, therefore, may be unsuitable for certain
a Numerical Aperture: Describes the angle in a cone of light emitted by the condenser and accepted by the objective of a microscope; the index of refraction of the medium in which the image lies multiplied by the sine of the half angle of the cone of light. b Rare gas: Alternative name for noble gas. Any of a group of six elements (helium, neon, argon, krypton, xenon, and radon), originally named ‘inert’ because they were thought not to enter into any chemical reactions.
Ch.1 Introduction 3
applications such as optical lithography despite its high peak power operation in the
VUV.
These rare-gas excimer lasers have been realized only by e-beam excitation until
recently [1-6]. Recently, the Kr2* excimer laser (the maximum output laser energy
was 150 µJ at 148 nm in 10 atm) have been realized by discharge-pumped scheme
[7]. This is a very interesting result. Despite various new schemes such as a four-
stage discharge [8], gas-jet discharge [5], silent discharge excitation [9] have been
studied and developed to obtain rare-gas excimer lasers action by discharge-pumped
scheme until recently, the Kr2* excimer laser action have been realized by a
conventional UV pre-ionized transverse compact discharge device finally. In the light
of this result, it is reasonable to suppose that there is the possibility of the Ar2*
excimer laser action by discharge-pumped method.
1.1 The electron beam pumped method
Figure 1.2 shows the fundamental electron beam-pumped laser components [10].
The electron beam source consists of a high voltage generator such as a Marx bank
or pulse transformer, a pulse transforming line to produce ideally square pulses of
20-100 ns duration and a vacuum diode. Electron emission is from a cold cathode
which is constructed from graphite or sharp blades to enhance the local electric fields
and produce more efficient and uniform emission. The anode consists of a foil,
usually of titanium, aluminum, stainless steel or aluminized dielectric, which is
sufficiently thin to allow efficient penetration by electrons with energies of 200 keV
or greater. The beams pass through a thin foil that isolates the diode from the high-
Ch.1 Introduction 4
pressure laser gas. The two vessels are separated by a thin foil the electrons have to
penetrate the foil which causes losses. The foil is a weak link in the system, reducing
the reliability of the device.
Fig. 1.2 The fundamental electron beam-pumped laser components [10]
1.2 The discharge pumped method
These types of discharge pumped lasers can be broadly classified into two
categories depending on the methods of generation of pre-ionization, viz., UV pre-
ionized lasers and X-ray pre-ionized lasers.
Figure 1.3 shows the fundamental X-ray pre-ionized laser components [11]. X-
ray photons have a high energy and a large penetration depth, thus resulting in a
Ch.1 Introduction 5
homogeneous electron density. However, X-rays are generated outside the laser
chamber and have much longer laser systems, albeit with substantial technological
complexity.
Fig 1.3 The fundamental X-ray pre-ionized laser components [11]
UV preionized discharge pumped lasers have the advantage of relative simplicity
and convenience. These types of UV pre-ionized lasers also, can be broadly
classified into two categories depending on the electrode structure for generation of
pre-ionization, viz., double discharge lasers and corona pre-ionized lasers.
In the double discharge laser, although simple unsustained discharge are
fundamentally unstable, fast, transverse discharge have been very successfully to
Ch.1 Introduction 6
pump lasers and have the advantage of simplicity of construction. At atmospheric
pressures and high electric fields, the electrons multiply very rapidly, the plasma
impedance collapses and spatial non-uniformities grow to form arcs in times which,
at pressure of several atmospherics, can be as short as a few tens of nanoseconds.
Figure 1.4 shows the fundamental double discharge transversely excited (TE) laser
components, the array of pre-ionizing pins was placed on one side at a distance of
~10 mm from the centre of the electrodes resulting in the shifting of the glow
discharge towards the pre-ionizer. When the main discharge was initiated after a
suitable time delay with respect to the pre-ionizing discharge, its stable self-sustained
avalanche mode of operation lasting several microseconds was possible resulting in
the efficient operation of a gas laser.
Heat Exchanger
Fan
Spark
Pin-electrode (for pre-ionization)
Electrode
Laser Beam
ElectrodeMain-discharge
Laser GasRecirculator
Mirror
Fig. 1.4 The fundamental double discharge TE laser components
Ch.1 Introduction 7
The second method of electron seeding the gas volume prior to the application of
the main discharge is to use the ultraviolet light by initiating a corona discharge. This
technique, a schematic diagram of which is shown in Fig. 1.5, was first demonstrated
by Lamberton and Pearson [12] in the operation of a TEA (Transversely Excited,
Atmospheric pressure) CO2 laser. The corona pre-ionized method is the high pre-
ionization uniformity, determined by a very homogeneous UV radiation flux that can
be produced along the main-discharge electrodes during the entire stage of
development of the main-discharge, with no jitter and very little delay.
H.V H.V
Cathode
Anode
Dielectric
(a) Plate type (b) Tube type
Fig. 1.5 Schematic diagrams of the corona pre-ionization
The double discharge method, which are particularly suitable for pumping very
large volume systems, are relatively cumbersome and cannot be easily scaled down
to small dimensions. Corona pre-ionized TE lasers, on the other hand, are easily
amenable to miniaturization.
Ch.1 Introduction 8
1.3 Aim of this study
The aim of this study is the investigation of the possibility of the Ar2* excimer
laser action and the development of the electrode structure for the efficient TE gas
laser. A new pre-ionization electrode is designed using the surface-corona pre-
ionization method. It is investigated experimentally in terms of generated charge
density, electrical characteristics. The automatically pre-ionized (API) plasma
electrode system also is designed using a new pre-ionization electrode by the
surface-corona pre-ionization method. Characteristics of the main-discharge in multi-
atmospheric (1~5 atm) pure Ar gas have been investigated using the automatically
pre-ionized (API) plasma electrode. It is examined the time dependent main-
discharge from two-dimensional simulation of electric field distribution of the API
plasma electrode discharge system. In addition, the light emission characteristics of
the discharge system using the API plasma electrode are investigated by the
measurement of Ar atomic line and Ar ionic line. On the basis of such experimental
results, it is the purpose to design the new resonator for the Ar2* excimer laser action.
Ch.1 Introduction 9
Reference
[1] W. M. Hugdes, J. Shannon, and C.K. Rhodes, 126.1-nm molecular argon laser, Appl. Phys. Lett. 24, 488 (1974) [2] W-G. Wrobel, H. Röhr, and K-H. Steuer, Tunable vacuum ultraviolet laser action by argon excimers, Appl. Phys. Lett. 36, 113 (1980) [3] Y. Uehara, W. Sasaki, S. Kasai, S. Saito, E. Fujiwara, Y. Kato, C. Yamanaka, M. Yamanaka, K. Tsuchida, and J. Fujita, Tunable oscillation of a high-power argon excimer laser in the vacuum-ultraviolet spectral region, Opt. Lett. 10, 487 (1985)
[4] K. Kurosawa, Y. Takigawa, W. Sasaki, M. Okuda, E. Fujiwara, K. Yoshida, and Y. Kato, High-power operation of an argon excimer laser with a MgF2 and SiC cavity, IEEE J. Quantum Electron. QE-27, 71 (1991) [5] S. K. Searles, J. E. Tucker, B. L. Wexler, and M. F. Masters, VUV emission from discharge-pumped Ar supersonic jets under high-excitation conditions, IEEE J. Quantum Electron. QE-30, 2141 (1994) [6] T Sakurai, N Goto, and C E Webb, Kr2* excimer emission from multi-atmosphere discharges in Kr, Kr-He and Kr-Ne mixtures, J. Phys. D: Appl. Phys. 20, 709 (1987) [7] W. Sasaki, T. Shirai, S. Kubodera, J. Kawanaka, and T. Igarashi, Observation of vacuum-ultraviolet Kr2* laser oscillation pumped by a compact discharge device, Opt. Lett 26, 503 (2001) [8] H. Ninomiya, and K. Nakamura, Ar2* excimer emission from a pulsed electric discharge in pure Ar gas, Opt. Commun. 134, 521 (1997)
[9] S. Kubodera, M. Kitahara, J. Kawanaka, W. Sasaki, and K. Kurosawa, A vacuum ultraviolet flash lamp with extremely broadened emission spectra, Appl. Phys. Lett. 69, 452 (1996)
Ch.1 Introduction 10
[10] J. D. Sethian, M. Friedman, J. L. Giuliani, R. H. Lehmberg, S. P. Obenschain, P. Kepple, M. Wolford, F. Hegeler, S. B. Swanekamp, D. Weidenheimer, D. Welch, D. V. Rose, and S. Searles, Electron beam pumped KrF lasers for fusion energy, Phys. Plasmas. 10, 2142 (2003)
[11] L. Feenstra, O. B. Hoekstra, P. J. M. Peters, and W. J. Witteman, On the performance of an ArF and a KrF laser as a function of the preionisation timing and the excitation mode, Appl. Phys. B 70, 213 (2000)
[12] P.R. Pearson and H. M. Lamberton, Atmospheric Pressure CO2 laser giving high output energy per unit volume, IEEE J. Quantum Electron. QE-8, 145 (1972)
Ch.2 Theoretical considerations 11
2. Theoretical considerations
The Ar2* excimer laser, in the case of electron-beam pumping, the pumping rate
exceeds 100 MW/cm3, which is quite high compared with that of rare-gas-halide
excimer lasers in the ultraviolet region and the F2 molecular laser at 157 nm, which
are typically of the order of less than 10 MW/cm3. In Table 2.1, the values of the
stimulated emissiona cross section (σ) times the lifetime (τ) of the laser upper state
(στ) are summarized. In the case of Ar2*, the value of στ is smaller almost by two
orders of magnitude, due to its smaller stimulated emission cross section caused by
its wide fluorescence spectrum. Since the minimum pump rate to reach the threshold
is basically proportional to 1/στ, the Ar2* excimer needs a higher pumping rate than
the other lasers listed above. It is very difficult to sustain a stable electrical discharge
during high-pressure Ar gas pumping with such a high power density as 100
MW/cm3 [1].b
Table 2.1 Comparison of spectroscopic parameters
Lifetime τ (ns)
Stimulated emission cross section σ (cm2 )
στ (s cm2)
KrF
F2
Ar2*
6.5
3.7
4.2
2 × 10−16
6.8 × 10−16
8 × 10−18
1.3 × 10−24
2.5 × 10−24
3.4 × 10−26
a Stimulated Emission: in a quantum mechanical system, the radiation emitted when the internal energy of the system drops from an excited level (induced by the presence of radiant energy at the same frequency) to a lower level. b It was explained in Appendix #A.
Ch.2 Theoretical considerations 12
2.1 Rare gas kinetics considerations
A schematic diagram is provided in Fig 2.1. The primary electrons deposit their
energy through ionization
e + X → X+ + 2e k1 (2.1)
The resulting secondary electrons cool through successive ionization steps and
atomic excitation
e + X → X* + 2e k2 (2.2)
until their energy drops below the excitation threshold. Subsequent elastic collisions
rapidly cool the electrons to a few tenths of 1 eV, where continued cooling is retarded
by the Ramsauer minimum in the scattering cross section.
The electron-ion recombination in these systems occurs mainly through
dissociative recombination with the molecular ion after it is formed by three-body
association:
X+ + 2X→ X2+ + X k3 (2.3)
X2+ + e → X** + X k4 (2.4)
Ch.2 Theoretical considerations 13
X (1S)
e
e
e
ee
e
e
e
e
e
X+ (1PJ)
ENERGYTRANSFER
ENERGYTRANSFER
X*
X**
X* (1,3PJ)2X
2X
X2**
X2*
2X
X2+ (2Σ)
X2* (1,3Σ)
X2 (1Σ)
hv
hv
Fig. 2.1 Schematic energy level diagram for rare gas excimers
This dissociative recombination predominantly populates a second group of
atomic exited states, X**[np5(n+1)p]. (The double asterisk indicates excitation above
the first excited levels.) The p levels rapidly relax via the reactions
X** + 2X→ X2** + X k5 (2.5)
X2** + (X) → X* + X + (X) k6 (2.6)
to the s levels that finally populate the excimer levels
X* + 2X → X2*(1,3Σg+ ) + X k7. (2.7)
In the absence of quenching by added gases, the excimer levels may decay through
Ch.2 Theoretical considerations 14
radiation:
X2* → 2X + hv k8. (2.8)
The fact that there are two excimer levels, the singleta and tripletb, greatly
complicates the interpretation of the fluorescence decay. The two excimer levels are
nearly coincident in energy, with the triplet lying less than 1000 cm-1 lower, but their
radiative lifetime differ greatly, as shown in Table 2.2. in the early afterglow, the
density of secondary electrons (at high pressures) can be sufficient to cause rapid
collision mixing of the tow levels, leading to a decay time representative of a
statistical 3-to-1 distribution of triplets over singlets. This leads to an affective
radiative lifetime approximately four times that of the singlet state (in the case Ar2*,
more five times), as shown in Table 2.2. In additional, the decay of the late afterglow
when the electron density is low is dominated by the triplets [2].
e + X2 (3Σu+ ) ↔ e + X2*(1Σu
+ ) k8. (2.9)
On the discharge pumped Ar gas excitation will be examined further in the 6 chapter.
a A singlet state is a state of an atom or molecule with zero net electronic spin(S=0). For two electrons, a singlet state is one with antiparallel (paired) spins, and is denoted ↑↓. b A triplet stats of an atom or molecule is a state in which the total spin quantum number is S=1. For two electrons, a triplet state corresponds parallel electron spins and is denoted ↑↑.
Ch.2 Theoretical considerations 15
Table 2.2 Radiative lifetimes of the excimer states
1Σu+ (nsec) 3Σu
+ (nsec) Statistical Average (nsec) Reference
Ar2*
Kr2*
Xe2*
4.2
3.3
4.8
2880
265
100
16.7
12.7
16.8
2.1
2.2
2.3
2.2 Discharge formation and stability in high pressure gas [3]
As well known in the gas discharge literature, there are two type of electrical
breakdown which can convert an initially nonconducting high-pressure gas between
two parallel electrodes into a highly conducting plasma upon the application of a
high-voltage pulse. One is the classical Townsend breakdown [4,5] and the other is
the plasma streamer breakdown [6]. Even though the basic process for electron
multiplication is due to electron avalanche in both types of breakdown, the
conditions for occurrence and applicable ranges of field strength to gas density ratio
E/n are quite different, so that it is important to make a distinction between the two.
The well-known Townsend or Paschen breakdown mechanism is characterized by a
large number of successive electron avalanches that originate from secondary
electron generation. The space-charge field caused by differential motions between
the electrons and the positive ions is assumed to be so weak as to be completely
negligible. Continuous exponentiation of the electron current within the discharge
gap is assumed to be maintained by a positive feedback of the Townsend avalanche
process through secondary electron emission at the cathode surface. A self-sustained
Ch.2 Theoretical considerations 16
discharge condition, corresponding to the onset of such positive feedback, is
accordingly given by an equation of the type
( ) ],/)1log[(/ γγα +=pdp (2.10)
where p is the gas pressure, d is the electrode gap, α is the first Townsend coefficient,
which measures the exponentiation rate of free electrons per unit mean drift distance
of the electrons under the influence of the constant applied electric field strength E
under consideration, and γ is the second Townsend coefficient, which measures the
total probability of secondary electron emission from all sources associated with a
single primary electron emission. If positive ion bombardment at the cathode surface
were the main source of secondary emission, the minimum time required for the
positive feedback mechanism to become effective after turning on the applied E field
at t=0 would then be some fraction of the ion transit time from anode to cathode,
,/ ii ud=τ (2.11)
where ui denotes the mean drift velocity of the positive ions. For He+ ions in Heat 1
atm pressure, ui is 5×102 m/sec at a typical breakdown field strength of 4×105 V/cm
[7]. Thus, for a single transit across a 4 cm electrode gap, τi ~10-4 sec. For heavier
ions across large gaps, the transit time would be correspondingly longer. On the other
hand, if photoelectric effects at the cathode were an important source of secondary
emission, the minimum time for positive feedback would then be governed either by
the characteristic time for generation of the appropriate excited molecular states
during the avalanche process or by the radiative lifetime of the excited molecule,
Ch.2 Theoretical considerations 17
whichever is longer. In any case, a relatively long formative time delay (~10-6 sec) is
generally observed in a Townsend-type breakdown which leads to sparking across
the gap.
Anode
Cathode
Anode
Cathode
++
++
++
+
+
+
+
(a) (b) (c)
Anode
Cathode
++
+
Fig. 2.2 Schematic diagrams showing (a) streamer development around a single primary electron avalanche after its space-charge field has grown beyond a certain critical value, (b) continuous backward propagation of the cathode-directed plasma streamer after the arrival of the primary avalanche head at the anode, and (c) complete bridging of the electrode gap by the plasma streamer
In contrast to the Townsend breakdown model, “Kanal” or streamer breakdown
occurs as the result of a large space-charge field that develops from a single electron
avalanche process into a rapidly propagating plasma streamer. This form of
breakdown will therefore allow the sparking phenomenon to begin anywhere inside
the discharge gap without relying on the secondary electron generation processes at
the cathode surface. The large space-charge fields that develop are due to the
Ch.2 Theoretical considerations 18
relatively low mobility of the positive ions as compared to that of the electrons. On
the time scale of interest in a typical short duration pulsed discharge, the electrons
are free to move toward the anode while the ions are essentially frozen in space. For
simplicity, the propagating avalanche head filled mostly with free electrons can be
idealized as a negatively charged sphere, behind which is the positive space-charge
[see Fig. 2.2(a)]. The shape of the avalanche cone is determined primarily by electron
diffusion [5]. At some critical point where the space-charge field of the avalanche
head becomes comparable in magnitude to that of the applied electric field E,
streamer development begins. At this point, secondary avalanches are initiated by
photo-ionization in front of and behind the head of the primary avalanche. Both an
anode- and cathode-directed streamer develop and move at a mush greater velocity
than the velocity of the avalanche head. The increased velocity of the anode-directed
steamer is due to the space-charge enhanced electric field on the anode side of the
avalanche head. The cathode-directed streamer is primarily the result of the positive
space charge left behind the avalanche head. In the surrounding gas, photoelectrons
are produced which initiate secondary avalanches directed along strong field lines
toward the stem of the primary avalanche. The greatest multiplication of these
secondary avalanches occurs along the axis of the primary avalanche where the
space-charge filed supplements the applied field. As the negatively charged
avalanche head propagates toward the anode and exponentiates, it also leaves behind
a positively-charged tail which continues to lengthen and intensify at an accelerating
pace until the anode and cathode are eventually connected by the self-propagating
plasma streamer. In Fig. 2.2, three successive stages of such streamer development
Ch.2 Theoretical considerations 19
are schematically illustrated. Thus, according to this model, breakdown will occur
whenever a single primary electron avalanche is allowed to develop to the critical
point of streamer initiation anywhere within the electron gap. A breakdown criterion,
attributed to Raether [8], which corresponds to the condition that the critical track
length ξc for a primary electron avalanche developed under the influence of a
suddenly applied constant electric filed is equal to the electrode gap d can
accordingly be derived, such that for air, in mks units,
( ) ).log(20/ dpdp +=α (2.12)
According to the early experiments of Townsend [4], the second ionization
coefficient γ is typically of the order of 0.1, so that the total avalanche gain αd
corresponding to the breakdown condition (2.10) is about 2.4, which is more than a
factor of 6 smaller than the value of αd corresponding to the breakdown condition
(2.12) for all electrode gaps of the order of a few centimeters or greater. Thus the
applied filed strength required for observation of a Townsend-type breakdown is
generally much weaker than that required for observation of a streamer breakdown.
On the other hand, for large-volume high-pressure discharges at high values of E/n
corresponding to those of general interest in high-power excimer laser excitation as
mentioned earlier, the total avalanche gain across the discharge gap often exceeds a
numerical value of 20 so that a streamer-type breakdown can be initiated from points
far away from the cathode due to the relative shortness of the critical avalanche track
length ξc in comparison with the electrode gap d. Furthermore, due to the nonlinear
buildup of the space-charge filed after the primary avalanche track length has grown
Ch.2 Theoretical considerations 20
beyond ξc within the electrode gap, a streamer-type breakdown can take place in a
time scale much shorter than the characteristic time
,/ ee ud=τ (4)
for the single transit of a primary electron from cathode to anode at a constant drift
velocity ue. For a fee electron in pure He or in a predominantly He gas mixture at 1
atm pressure under the influence of a typical breakdown filed strength E=4×105 V/m
[3], ue is about 2×104 m/sec, so that τe ~ 2×10-6 sec for a 4 cm electrode gap. This
explains why streamer breakdown has been observed with formative delay times as
shot as 10-9 sec in some fast pulse discharge at high values of E/n.
For large scale homogeneous discharge it is necessary to pre-ionize the gas so as
to form a homogeneous electron density distribution, from which a large number of
electron avalanches can start simultaneously. The resulting overlapping electron
clouds and ion-cones form homogenous plasma (see Fig 2.3).
Ch.2 Theoretical considerations 21
Fig. 2.3 The effect of a homogeneous pre-ionization electron density: the electron clouds and the ion-cones overlap, yielding a homogeneous charge carrier density and thus homogeneous discharge plasma
Several authors have discussed the theoretical criteria for a minimum pre-
ionization electron density to ensure the good discharge homogeneity, e.g. see [3, 9].
In the calculations of Levatter and Lin, ref. [3], an average minimum pre-ionization
electron density of approximately 108 cm-3 and with a long voltage rise time a lower
electron density can be allowed. In order to start a discharge, the pre-ionization
Ch.2 Theoretical considerations 22
electron density must be multiplied to the discharge level of ~1014 cm-3 by the
electric filed.
2.3 General characteristics of the glow (or uniform) discharge
If a glow discharge is maintained between two electrodes in a long glass tube,
several bright and dark regions will be observed as shown in Fig. 2.4(a). The
variation of the relative intensity of these regions is shown Fig. 2.4(d). It can be seen
that (i) the negative glow region is the brightest region of the glow discharge; (ii) the
positive column constitutes a uniform and large bright region; and (iii) the anode and
the Faraday dark spaces are not completely non-luminous and they are usually
difficult to observe experimentally. The dark region that is clearly observed
experimentally is the layer between the cathode surface and the negative glow which
is called the cathode-fall region. This region is of interest due to the fact that most of
the voltage maintained between the electrodes is dropped across this region. The
distribution of the potential along the tube is shown in Fig. 2.4(c). The potential rises
rapidly from zero at the cathode and reaches a value of cathode voltage Vc at the
edge of the negative glow. From this point onwards, the potential stays
approximately constant, or rises very gradually. The rise becomes slightly sharper
near the anode.
Ch.2 Theoretical considerations 23
Fig. 2.4 Characteristics of a typical glow discharge
Ch.2 Theoretical considerations 24
The physical appearance of the glow discharge has been found to be affected
by system parameters such as the gas pressure; the electrode separation; the type of
gas used; the current that flows and the cathode material. Their effects are described
briefly as follows:
a) Effect of the gas pressure
The pressure of the gas in the discharge tube has a significant effect on the
relative length of the various regions of the glow discharge. Generally, the glow
discharge is operated at a pressure of less than 100 torr. At the high pressure and, the
positive column is favoured. The negative glow and the cathode-fall region are
compressed at high pressure while positive column extends to fill the space between
the electrodes. However its radial dimension also shrinks and it becomes not in
contact with the wall of the glass tube anymore. On the other hand, low pressure
operation of the glow discharge results in the shrinking of the positive column in the
axial direction while the cathode-fall region and the negative glow extend to fill the
tube. At very low pressure, probably in the region of milli-torr, the positive column
may disappear completely.
b) Effect of electrode separation
Once a glow discharge has been initiated, the voltage required to maintain it
will only increase slightly if the distance between the electrodes is increased. The
main effect of increasing the electrode separation is on the length of the positive
column – it increases to fill the extra space created by increasing the separation while
the other regions remain practically undisturbed.
Ch.2 Theoretical considerations 25
c) Effect of the type of gas used
The general description of the appearance of the glow discharge is valid for any type
of gas used. However, the colours of the luminous regions may vary for different
type of gas used. The characteristic colours for some commonly used gases are listed
in Table 2.3 below [10];
Table 2.3 Characteristic colours of glow discharge plasmas
Gas Cathode layer Negative glow Positive column
Ne
Air
H2
Yellow
Pink
Brownish-red
Orange
Blue
Pale blue
Brick red
Red
Pink
O2 Red Yellowish-white Pale yellow with
pink center
N2
Ar
He
Pink
Pink
Red
Blue
Dark blue
Green
Red
Dark red
Red to violet
Ch.2 Theoretical considerations 26
2.4 Excitation circuits
H.V.Cm
Charging bypass (R or L)
Swich
V0
t0
-V0
-2V0
H.V.Cm
Charging bypass (R or L)
CpSwich
V0
(a) LC Circuit or RC Circuit
(a) and (b) Discharge Waveform
(b) Charge Transfer type Circuit
Fig 2.5 Typical excitation circuits and discharge waveform for a TE laser: (a)LC circuit or RC circuit and (b) Charge transfer type circuit
Typical excitation circuits and discharge waveforms for a TE laser are shown
in Figs. 2.5 and 2.6. In the case of an excitation circuit a Fig. 2.5(a), DC high voltage
supply charges up a condenser through a charging element, normally a resistance or
an inductance. The charging bypass provides a path for the charging current. Once
the condenser is charged to the required voltage, the rapid closure of the high voltage
and high current switch enables the condenser to deliver its stored energy into the
laser load before glow-to-arc transition can occur. The charging bypass must offer
impedance that is many times more than that of the laser load lest this should eat up a
significant fraction of the energy stored in the condenser lowering, thereby, the plug
Ch.2 Theoretical considerations 27
in efficiency of the laser. At the same time its impedance should be much less than
that of the charging element so that the current flowing through the conducting
switch from the source following a discharge can be kept low for a given repetition
rate. In the single shot operation, the condenser is normally charged resistively and a
spark gap is traditionally used as a switch. For repetitive operation, however, more
efficient charging by means of inductance is employed and a thyratron replaces the
spark gap. In the case of an excitation circuit a Fig. 2.5(b), the capacitance Cm is
charged to ~10 kV and on closing the switch which can be either a spark gap or
thyratron, the voltage on Cp and therefore between the discharge electrodes rises
rapidly and breakdown occurs in the laser cavity. It is important to minimize the
inductance of the loop compressing Cp and the laser cavity and hence Cp generally
consists of a large number of low inductance capacitors distributed along and close to
the discharge channel.
Figure 2.6 shows several excitation circuits and discharge waveforms for a
TE laser: (a) LC inversion circuit (b) pulse forming circuit and (c) Blumlein circuit.
A circuit (a) can be used to supply a more high voltage across the load. Circuit (b)
and (c) can be used to supply a fast-rising voltage across the load.
Ch.2 Theoretical considerations 28
H.V. Swich
Pulse Forming Line(PFL)
t0
-V0
-2V0
t0
-V0
-2V0H.V.
Charging bypass
(L)
Swich
V0
C1
C2
H.V.
Swich Blumlein
t0
-V0
-2V0
(a) LC inversion Circuit (a) Discharge Waveform
(b) Pulse Forming Circuit (b) Discharge Waveform
(c) Blumlein Circuit (c) Discharge Waveform
Fig 2.6 Several excitation circuits and discharge waveforms for a TE laser: (a) LC inversion circuit (b) pulse forming circuit and (c) Blumlein circuit
2.5 The main discharge
During stable period of the discharge a glow discharge is formed. A general
property of a glow discharge is its constant voltage across the electrode almost
independent of the current flowing through the discharge. At this voltage there is
Ch.2 Theoretical considerations 29
equilibrium between the production and the loss of electrons. If the discharge
becomes unstable during the current pulse, i.e. if this equilibrium is disturbed locally,
streamers and eventually arcs will start to glow. Arcs shorten the life time of the
electrodes, so to prevent electrode wear the system has been optimized to prevent
arcs.
L
Cp
Vp Vss
Discharge
i
Fig 2.7 Simple LC model for the discharge circuit during steady state
During the steady state phase of the discharge, the electrical circuit can be described
with a simple LC-circuit as shown in Fig. 2.7. This simple circuit can only be used
for the first half cycle of the current. The voltage across the electrodes is kept
constant at the steady state voltage Vss. The differential equation of this circuit is
02
2
=+dt
idLCpi
(2.5)
with the initial conditions
Ch.2 Theoretical considerations 30
LVV
ii ssp −=′= )0(0)0( (2.6)
The solution of this differential equation and its starting conditions is
)sin()()(LCptVV
LCpti ssp −= (2.7)
In these equations Vp is the charging voltage. Cp and L are the capacitance of the
capacitor bank and inductance of the circuit respectively.
Ch.2 Theoretical considerations 31
Reference [1] H. Tanaka1, A. Takahashi1, T. Okada1, M. Maeda1, K. uchino1, T. Nishisaka, A. Sumitani, and H. Mizoguchi, Production of laser-heated plasma in high-pressure Ar gas and emission characteristics of vacuum ultraviolet radiation from Ar2 excimers, Appl. Phys. B 74, 323 (2002) [2] E. Zamir, D. Huestis, H, Nakano, R. Hill, and D. Lorents, Visible absorption by electron-beam pumped rare gases, IEEE J. Quantum Electron. QE-15, 281 (1979) [3] I. Jeffrey, Levatter, and S. C. Lin, Necessary conditions for the homogeneous formation of pulsed avalanche discharges at high gas pressures, J. Appl. Phys. 51, 210 (1980) [4] J. S. Townsend, Electricity in Gases, Oxford: Oxford University Press (1915) [5] H. S. W. Massey, Electronic and Ionic Impact Phenomena. Volume II: Electron Collisions With Molecules and Photo-Ionization, 2nd ed. Oxford: Oxford University Press (1969) [6] J. A. Rees, Electrical Breakdown in Gases, London: Macmillan (1973) [7] A. V. Phelps and S. C. Brown, Positive Ions in the Afterglow of a Low Pressure Helium Discharge, Phys. Rev. 86, 102 (1952) [8] H. Reather, Arch. Electrotech. (Berlin) 34, 49 (1940) [9] A. J. Palmer, A physical model on the initiation of the atmospheric-pressure glow discharges, Appl. Phys. Lett. 25, 138 (1974) [10] S.C. Brown, Introduction to Electrical Discharge in Gases, U.S.A: John Wiley (1966)
Ch.3 The pre-ionization source for excimer laser 32
3. The pre-ionization source for excimer laser
It is well known that an efficient laser excitation in pulsed transversely excited
(TE) gas lasers occurs in the glow discharges mode. The formation of streamers in
self-sustained discharges and their subsequent degeneration into arcs can be avoided
by pre-ionizing the discharge volume. The quality of the pre-ionization in terms of
density and uniformity of the produced photoelectrons is important because it
directly affects on the laser performance characteristics. By improving the pre-
ionization one can extend the laser operating conditions and increase the input energy
that can be deposited into the active medium while still maintaining a homogeneous
discharge. A variety of pre-ionization methods are being used [1,2], generally based
upon the generation of UV photons by discharge which are fired in advance of the
main excitation discharges. Among these methods, one of the most effective method
relies on corona discharges over the dielectric surface as UV radiation sources [3,4].
Pre-ionizing schemes which utilize this technique have proven to originate very
durable and reliable devices [5,6], and provide a viable means for reducing size, cost,
and complexity of TE gas lasers.
A main feature of the corona-discharge pre-ionization schemes is the high pre-
ionization with relative uniformly, determined by a very homogeneous UV radiation
flux that can be produced along the main discharge electrodes during the entire stage
of development of the main discharge, with no jitter and very little delay. Therefore,
to realize higher stability and efficiency of TE gas laser, we have been designed a
new pre-ionization electrode (see Fig. 3.1) for the generation of UV radiation from a
Ch.3 The pre-ionization source for excimer laser 33
surface-corona discharge over the surface of a dielectric.
3.1 Experimental set-up and methods
Figure 3.1 shows a schematic diagram of surface-corona electrode for the
pre-ionization. A surface-corona electrode was consisted with a stick-shaped inner
conductor of 9.5 mm in diameter as a high potential electrode, which inserted in the
cylinder-shaped ceramic (Al2O3) tube of 10 mm in inside diameter and 13.4 mm in
outside diameter, and a 1/4 cylinder-shaped copper board as a ground electrode.
Figure 3.2 show an experimental set-up and an equivalent circuit. In this circuit, DC
high voltage supply charges up a condenser through a charging element, normally a
resistance or an inductance. The charging bypass provides a path for the charging
current. Once the condenser is charged to the required voltage, the rapid closure of
the high voltage and high current switch enables the condenser to deliver its stored
energy into the surface-corona electrode.
Ch.3 The pre-ionization source for excimer laser 34
Rod electrode(H.V.)
Earth electrode
Ceramic tube
Surface-corona discharge
Fig. 3.1 Schematic diagram of a surface-corona electrode for the pre-ionization
Fig. 3.2 Experimental set-up and an equivalent circuit
A delayed pulse generator was used to adjust the timing between the surface-
corona discharge and the triggering of the ICCD camera. It was set that the operation
gas was Ne, Ar and F in 1000 Torr, applied voltage of 10 kV, and exposure time in 50
ns for the ICCD (Image Intensifier Charge-Coupled Device) camera. Special
attention was paid to charge the corona electrode set position, which can be described
Ch.3 The pre-ionization source for excimer laser 35
by the deviation angle against the center of the slit to the edge of the 1/4 cylinder-
shaped copper broad electrode, and makes a minus for the clockwise, and counter
clockwise is a plus. The corrector electrodes shown in Fig. 3.2 measured charged
particle densities against the deviation angle [7].
3.2 Experimental results and discussions
Many pictures were taken by the ICCD camera to investigate the discharge
development with different delay time. Figure 3.3 shows the spatial-temporal
evolution of corona-surface discharge in Ne gas. Figure 3.4 shows the spatial-
temporal evolution of corona-surface discharge: (a) Ne gas and (b) mixed gas (Ne:
97.8 %, Ar: 2 %, F2: 0.2 %) at atmospheric pressure respectively. It knows that
corona discharge starts from the edge of the ground electrode and develops on the
surface of ceramic tube. Because a minus high voltage was applied to the inner
electrode, it should presume a positive streamer discharge, which crawled on the
surface of the ceramic tube. Moreover, both timing of the occurrence and the
development of the corona discharge were about simultaneous due to the uniformity
of pre-discharge along the axis direction of ceramic tube. Considering Thyratron
jitters in 10-20 ns, it was found that the development speed of the surface-corona
discharge was 107-108 cm/s in pure Ne gas.
Ch.3 The pre-ionization source for excimer laser 36
Fig. 3.3 Spatial-temporal evolution of corona-surface discharge in Ne gas
Earth electrode
Observation region using an ICCD camera
Earth electrode
Time
(a)
10 ns 60 ns 130 ns
10 mm
(b)
30 mm
H.V
Fig. 3.4 Spatial-temporal evolution of corona-surface discharge: (a) Ne gas and (b) mixed gas (Ne: 97.8 %, Ar: 2 %, F2: 0.2 %)
Ch.3 The pre-ionization source for excimer laser 37
ICCDCamera
-xO
Ceramic tube
Earth electrode
-60 -50 -40 -30 -20 -10 0
4
6
8
10
12
14
16
18
20
22
24
Den
sity
of c
harg
ed p
artic
le (
107 /c
m3 )
Angle ( xO)
8kV 10kV 12kV
Fig. 3.5 Charged particle density against the deviation angle in Ne gas
Figure 3.5 shows the charged particle density against the deviation angle.
The results show that the highest density of charged particles is obtained at the -30
degree deviation angle and density of charged particles increases with increasing
applied voltage. Therefore, a sharp edge of the ground electrode is possible to radiate
the strong UV light.
Ch.3 The pre-ionization source for excimer laser 38
Reference
[1] M. Richardson, K. Leopold, and A. Alcock, Large aperture CO2 laser discharges, IEEE J. Quantum Electron. QE-9, 934 (1973) [2] V. Hasson and H. M. von Bergmann, Ultraminiature high-power gas discharge lasers, Rev. Sci. Instrum. 50, 59 (1979) [3] G. J. Ernst and A. G. Boer, Construction and performance characteristics of a rapid discharge TEA CO2 laser, Opt. Commun. 27, 105 (1978) [4] V. Hasson and H. M. von Bergmann, Simple and compact photopreionization-stabilized excimer lasers, Rev. Sci. Instrum. 50, 1542 (1979) [5] R. Sze, and E. Seegmiller, Operating characteristics of a high repetition rate miniature rare-gas halide laser, IEEE J. Quantum Electron. QE-17, 81 (1981) [6] R. Marchetti, E. Penco, and G. Salvetti, Compact sealed TEA CO2lasers with corona-discharge preionization, IEEE J. Quantum Electron. QE-19, 1488 (1983) [7] K. Fukuda, N. Hayashi, S. Satoh and C. Yamabe, Estimation of Corona Preionization for Excimer Laser, Rep. Fac. Eng. Saga Univ., 29, 35 (2000) (in Japanese)
Ch.4 The API plasma electrode discharge system 39
4. The Automatically Pre-Ionized plasma electrode
discharge system
The questions surrounding discharge stability became particularly relevant with
the advent of high-pressure electrical-discharge lasers in the 1970s, and a
considerable number of studies have been conducted on this topic. The mechanisms
of spark formation have been established [1-3]. The main factors limiting maximum
energy deposition are the formation of cathode spots during field emission, and
explosive processes occurring at the cathode in the strong field of the cathode
potential fall. As a result of electron emission fluctuations from the cathode, a highly
conductive channel forms near one of the spots, and the current from the entire
cathode surface is drawn into this channel. A plasma electrode has been developed to
control such cathode spots [4,5]. The disadvantages of such plasma electrodes, as
reported in the literature [6], include the deflection of the main-discharge and the
need of an auxiliary circuit to form the surface discharges channel. Therefore, the
API (automatically pre-ionized) transversely excited plasma electrode was designed
to solve these problems. The structure of this API plasma electrode has been derived
from the “double discharge” electrode structure [7,8].
4.1 Experimental set-up
Figure 4.1 shows a schematic diagram of a cross-sectional view of the API
plasma electrode discharge system used in this work. Figure 4.2 shows the structure
Ch.4 The API plasma electrode discharge system 40
of the API plasma electrode. The surface-discharge plasma that forms on the
dielectric surface provides intense ultra-violet radiation, suitable for pre-ionization in
the main-discharge space and able to sustain a surface conductivity high enough to
serve as an electrode.
Fig. 4.1 Schematic diagram of a cross-sectional view of the API plasma electrode discharge system
Ch.4 The API plasma electrode discharge system 41
Fig. 4.2 Structure of the API plasma electrode
Therefore, compared with the conventional flat-plate electrode, the API
electrode can supply relatively high-energy in the main-discharge region and can
persist for a long time. The discharge electrode consists of a 2 mm thick ceramic
(Al2O3: purity 99%) tube whose almost half of outside is covered with aluminum
plates and is clogged inside with an aluminum rod. The surface discharge area of the
API plasma electrode is 2.2 cm × 74 cm. The discharge gap between electrodes is 0.3
cm. As shown in Fig. 4.3, the equivalent circuit was the conventional charge transfer
type [9,10], but for two other points, i.e. the dielectric capacitance was varied by
using a special electrode structure, and the operation was performed with and without
capacitance, Cpr, which controls the pre-ionization. The circuit parameters indicated
in Fig. 4.3 and Table. 4.1 were estimated from the measurements of temporal
changes of the current which passes from Cm to Cp with a Rogowski coil and the
voltage applied across the electrodes was measured with a high-voltage probe (P6015,
Tektronix). Figure 4.4 shows a schematic describing observation of the discharge
Capacitors for pre-ionization, Cpr (on anode)
Peaking capacitance, Cp Dielectric (Al2O3 99.0%)
Cathode
Anode
Capacitors for pre-ionization, Cpr (on cathode) Aluminum Rogowski Coil
Ch.4 The API plasma electrode discharge system 42
formation properties of the main-discharge. An ICCD camera with high-speed
electronic shutter (C5909, Hamamatsu, Japan) was used to observe the discharge of
~10-ns order. To observe the discharge form, quartz was used as the window material,
but an optical resonator was not used. In order to measure the emission characteristic
from the Ar discharge, a grating monochromator (CT-25C, JASCO) and a
photomultiplier tube (R372, Hamamatsu, Japan) were used. The gas used for these
measurements was pure Ar gas at longer than one atmospheric pressure. The
experiments were performed at a repetition rate of 1 Hz.
Fig. 4.3 Equivalent electrical circuit of the API plasma electrode discharge system
Ch.4 The API plasma electrode discharge system 43
Cathode
Anode
CHAMBER
Power Circuit
Powercontroller
ICCDCamera
ICCD Controller
Filter
DELAYGENERATOR
DG 535
Fig. 4.4 Schematic for the observation of the discharge formation characteristics
4.2 Circuit characteristics
The driving parameters of the API plasma electrode discharge system are
shown in table.1. Behaviors of the API plasma electrode discharge system and the
conventional charge transfer system are almost same as shown in Fig. 4.3. Figure 4.5
shows the typical signal waveforms using the automatically pre-ionized charge
transfer system with the auxiliary pin electrode. The main capacitance Cm is charged
by the required voltage V0. As shown in Fig. 4.3, during the S/W closed, the electric
energy Ep = (1/2) CpVbr2 is transferred from Cm to Cp and also, the electric energy Epr
= (1/2) [(Cpr Cdi)/ (Cpr+ Cdi)]Vbr2 transferred from Cm to Cpr at the same time, where
Vbr is a breakdown voltage and Cdi is the dielectric capacitance between an inside
aluminum rod electrode(covered with Al2O3) and an outside aluminum electrode.
Ch.4 The API plasma electrode discharge system 44
Table 4.1 Experimental results obtained by the API plasma electrode discharge system shown in Fig. 4.6
Parameter
Vo
Vbr
S/W
Rs
Rch
Cm
Cp
Cpr
Cdi
Lm
Lp
: Initial store voltage 20 kV
: Breakdown voltage
: Spark gap switch
: Resistance of the spark-gap S/W 0.5 Ω
: Charging bypass resistance 2.5 kΩ
: Main capacitance for energy store 36 nF
: Peaking capacitance 29.4 nF
: Capacitance for pre-ionization and surface plasma 350 pF
: Capacitance of ceramic tube for pre-discharge
: Circuit inductance for the current flow from Cm to Cp ~100nH
: Self-inductance of the main-discharge ~10nH
Electrode separation 0.3 cm
The Cpr is used to control the effect of the capacitance of ceramic tube Cdi on
the discharge system of the API plasma electrode and also for prevention from the
electrical breakdown of ceramic tube. Before the main-discharge is formed, the
electric energy Epr produces surface-corona discharge on the cathode surface, and
then the auxiliary plasma channel is generated. Consequently, the electric energy Ep
forms the main-discharge in the discharge space by the auxiliary plasma channel.
Ch.4 The API plasma electrode discharge system 45
Initial discharge voltage VbrCm Cp Charging time
0Voltage
Time
Switch On
Pre-ionized discharge
Current
Fig. 4.5 Typical signal waveforms using the automatically pre-ionized charge transfer system with the auxiliary pin electrode (see Fig. 1.4)
4.3 Experimental results and discussions
4.3.1 Discharge characteristic of the API plasma electrode
Figure 4.6 shows that the variation of the main-discharge with Ar gas
pressure in the discharge system of the API plasma electrode. The main-discharge
volume is increased with increasing Ar gas pressure, and the form of the main-
discharge volume is distributed very uniformly but it is determined by the extent of
the surface-corona discharge.
Ch.4 The API plasma electrode discharge system 46
1 atm 2 atm 3 atm 4 atm 5 atmWindowDielectricMetal
Main-discharge Fig. 4.6 Variation of the main-discharge with Ar gas pressure in the discharge system of the API plasma electrode (at the formed Vbr, V0 = 20 kV)
The latter is limited by the value of the voltage required for ignition of the
surface-corona discharge. The breakdown voltage Vbr increased with increasing Ar
gas pressure as shown in table. 4.2. It can be considered that the impedance of
discharge space increases with increasing Ar gas pressure. In addition, the increase of
impedance leads to the increase of the discharge power by improvement of an
impedance matching as shown the variation of the discharge power with Ar gas
pressure in Fig. 4.8. The peak power put into the main discharge is about 90 MW.
Figure 4.7 shows the variation of the discharge voltage and current at 5 atm Ar gas.
The oscillation of discharge current (dotted line circle) is observed before the main-
discharge is formed. This oscillation seems to be due to the surface-corona discharge
that is the pre-ionization. In addition, the main-discharge forms according to the
density and distribution of electric charges created by the surface-corona discharge.
Unfortunately, we are unable to observe the longitudinal uniformity of the main-
discharge, because copper plates were used to connect parallel capacitors of API
plasma electrode discharge system shown in Fig. 4.2. However, longitudinal
Ch.4 The API plasma electrode discharge system 47
uniformity of the surface-corona discharge for the pre-ionization refers to a chapter 3
mentioned previously. Therefore for the experimental results shown in Fig. 4.6, it is
considered that the uniform main-discharge volume is formed because the surface-
corona discharge is formed uniformly.
Table 4.2 Experimental results obtained by the API plasma electrode discharge system shown in Fig. 4.6
1atm 2atm 3atm 4atm 5atm
Main-discharge volume [cm3] 13.7 38.9 47.8 62.8 67.4 Breakdown voltage Vbr [kV] 14.8 15.3 17.5 18.2 20
-40
-30
-20
-10
0
10
0 100 200 300 400 500 600 700 800 900 1000time[ns]
Dis
char
ge V
olta
ge [k
V]
-5
0
5
10
15
20
Dis
char
ge C
urre
nt [k
A]
Fig. 4.7 Variation of the discharge voltage and current with time at 5atm Ar gas (experiment of Fig. 4.3)
Ch.4 The API plasma electrode discharge system 48
Fig. 4.8 Variation of the discharge power with Ar gas pressure (experiment of Fig. 4.3)
4.3.2 Discharge characteristic of the API discharge system without Cpr
To observe the control effect of the surface-corona discharge (i.e. the pre-
ionization), the capacitance Cpr has been removed from the circuit the main-
discharge for different Ar pressure without Cpr is shown in Fig. 4.9. All experimental
conditions except the capacitance Cpr were same shown in Fig. 4.3. Comparing with
Fig. 4.6, the main-discharge volume in Fig. 4.9 deflected perceptibly to right side.
1 atm2 atm
3 atm
4 atm 5 atm
Ch.4 The API plasma electrode discharge system 49
Fig. 4.9 Main-discharge of API discharge system without Cpr for different Ar pressure
-40
-30
-20
-10
0
10
0 100 200 300 400 500 600 700 800 900 1000time[ns]
Dis
char
ge V
olta
ge [k
V]
-5
0
5
10
15
20
Dis
char
ge C
urre
nt [k
A]
Fig. 4.10 Variation of discharge voltage and current in the main-discharge with Ar gas pressure at API discharge system without Cpr (at the formed Vbr, V0 = 20 kV)
In addition, the oscillation of discharge current (dotted line circle in Fig.
4.10) is not observed before the main-discharge is formed. The main-discharge
volume was increased and the breakdown voltage Vbr was decreased, in comparison
with above experimental results. These results are due to the surface-corona
Ch.4 The API plasma electrode discharge system 50
discharge which is the pre-ionization was not formed uniformly. The applied voltage
on dielectric tube (capacitance Cdi) is approximately Vbr/2 when Cpr is operated, but
when there is no Cpr, the capacitor Cdi is applied Vbr. The cathode spots are formed in
the edge of metal electrode if the very high-voltage is applied at Cdi. As a result of
fluctuations of the electron emission from the cathode, a highly conductive channel is
formed near one of the spots, and the current from the entire cathode surface is
drawn into this channel. This phenomenon appears in Fig. 4.9. Consequently, when
compared with Fig. 4.8, the maximum discharge power in Fig. 4.11 was decreased
due to these cathode spots which is the main factor of limitation for the maximum
energy deposition. Therefore, according to the comparison of above two
experimental results, it is considered that the uniform pre-ionization forms the
uniform main-discharge by the control effect of Cpr, and the maximum energy
deposition is also increased.
Table 4.3 Experimental results obtained by API plasma electrode discharge system without Cpr shown in Fig. 4.9
1atm 2atm 3atm 4atm 5atm
Main-discharge volume [cm3] 39.6 41.6 73.7 81.5 92.6
Breakdown voltage Vbr [kV] 11.9 15.1 15.4 15.8 18.5
Ch.4 The API plasma electrode discharge system 51
Fig. 4.11 Variation of the discharge power with Ar gas pressure (experiment of Fig. 9)
1 atm
2 atm3 atm
4 atm 5 atm
Ch.4 The API plasma electrode discharge system 52
4.3.3 Using the plate electrode as an anode
In order to investigate the effect of different electrode configuration used
such as the plate electrode as an anode, we performed the experiment using a circuit
shown in Fig. 4.12. The electrode material of the anode was used copper and
experimental conditions except the discharge gap (0.6 cm) were same as these
denoted above. This experimental result was found that the main-discharge was
formed splitting by two as shown in Fig. 4.13, and main-discharge was transferred to
the arc-discharge after several ten ns. The discharge current could not be measured
due to the arc discharge. Therefore, the electrode structure of Fig. 4.13 is not suitable
for the formation of the uniform main-discharge. Although the shape of the discharge
electrode and the length of the discharge gap are different in this experiment, it is
inferred from above experimental results that the capacitance Cpr on the anode was
required as well as the cathode plasma for the formation of the uniform main-
discharge.
Fig. 4.12 Equivalent electrical circuit of discharge system used copper electrode on an anode
Ch.4 The API plasma electrode discharge system 53
Fig. 4.13 Variation of the main-discharge with Ar gas pressure of discharge system used copper electrode as an anode. (at the formed Vbr, V0 = 20 kV)
A copper electrode
After 70 ns
Clear arc discharge
5 atm 1 atm (b)1 atm (a)
Ch.4 The API plasma electrode discharge system 54
Reference [1] T. E. Broadbent, The breakdown mechanism of certain triggered spark gaps, Br. J. Appl. Phys. 8, 37 (1957) [2] P. W. Chan, R. J. Churchill, and M. S. Gautam, Radial recovery of high current spark channels, Int. J. Electronics. 32, 745 (1973) [3] I. Jeffrey, Levatter, and S. C. Lin, Necessary conditions for the homogeneous formation of pulsed avalanche discharges at high gas pressures, J. Appl. Phys. 51, 210 (1980) [4] K. Nakamura, N. Yukawa, T. Mochizuki, S. Horiguchi, and T. Nakaya, Optimization of the discharge characteristics of a laser device employing a plasma electrode, Appl. Phys. Lett. 49, 1493 (1986) [5] A. R. Sorokin and V. N. Ishchenko, High-power discharge with a plasma cathode in dense gases, Tech. Phys. 42, 1249 (1997) [6] V. Yu. Baranov, V. M. Borisov, A. M. Davidovskii and O. B. Khristoforov, Sov. J. Quantum Electron. 11, 42 (1981) [7] Yu Li Pan, A.T. Bernhardt and J. R. Simpson, Construction and Operation of a Double-Discharge TEA CO2 Laser, Rev. Sci. Instrum. 43, 662 (1972) [8] R. V. Bobcock, I. Lberman and W. D. Partlow, Volume ultraviolet preionization from bare sparks, IEEE J. Quantum Electron. 12, 29 (1976) [9] A. J. Andrews, A. J. Kearsley, C. E. Webb and S. C. Haydon, A KrF fast discharge laser in mixtures containing NF3, N2F4 or SF6, Opt. Commun. 20, 265 (1977) [10] R. C. Sze and P. B. Scott, 1/4-J discharge pumped KrF laser, Rev. Sci. Instrum. 49, 772 (1978)
Ch.5 The electric field of API plasma electrode system 55
5. The electric field of API plasma electrode system
In comparison with the results of the prior experiments with and without Cpr,
the main-discharge volume was increased and the breakdown voltage Vbr was
decreased for without Cpr. These changes are due to the surface-corona discharge not
forming uniformly. Considering these results in terms of the equivalent circuit, when
Cpr is operating, the voltage applied across the dielectric capacitor Cdi is
approximately Vbr/2, but without Cpr operating, the voltage applied across Cdi is Vbr.
If a very high-voltage is applied at Cdi, the cathode spots are formed at the edge of
the metal electrode. As a result of fluctuations in electron emission from the cathode,
a highly conductive channel is formed near one of these spots, and the current from
the entire cathode surface is drawn into this channel. This is the phenomenon that
appears in Fig. 4.9. Consequently, when compared with Fig. 4.8, the maximum
discharge power in Fig. 4.11 was decreased due to these cathode spots, which is the
main factor of the limitation for the maximum energy deposition. Therefore, from the
above two experimental results, it is considered that the reduction of discharge power
is caused by the non-uniform main-discharge. In addition, the non-uniform main-
discharge exists over an expanded region. This can be clarified by measurement of
the electric field distribution in the discharge region.
However, precise measurement of the electric field in the discharge region suffered
from technical limitations. As such, calculation of the potential distribution was
performed using the finite element method (FEM) in order to determine the electric
Ch.5 The electric field of API plasma electrode system 56
field variation. We used the general Galerkin finite-element software FlexPDETM [1],
which provides great flexibility in boundary geometry and formulation of equations.
Figure 5.1 shows the FEM modeling of the API plasma electrode.
Fig. 5.1 Modeling of the FEM for the API plasma electrode
In order to determine the time dependent of the electrical field in the API
plasma electrode, variable boundary conditions were obtained from actual
experimental data (the discharge voltage data at 5 atm) for the electrode with Cpr and
without Cpr, respectively. For example, variable boundary conditions in Fig. 5.1 are
applied to a aluminum Al_b, Al_t, Al_rod_b and Al_rod_t. In the case with Cpr, actual
experimental data (the time dependent breakdown voltage Vbr at 5 atm with Cpr:
Ch.5 The electric field of API plasma electrode system 57
sampling 2.0 GS/s) is applied to boundary condition Al_b and Al_rod_t is applied a
half of Vbr. Figures 5.2 ~ 5.7 show the electric field distribution (log scale) and
potential distribution lines at 50 ns ~ 350 ns after initial discharge (with Cpr and
without Cpr). Figures 5.3(a) and 5.6(a) show the calculated electric field distribution
(log scale) at 150 ns after initial discharge.
Ch.5 The electric field of API plasma electrode system 58
with cpr1: Cycle=1 Time= 50.000 dt= 50.000 p2 Nodes=15863 Cells=7656 RMS Err= 0.0323
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
a
a
a
a
m
m
n
o
o
op p
q
q q
q
r
r
r
r rrr
r
r r
r
r
s
s
ss
s
s
ss
ss
ss
ss
ss
t
t
tt t
u
uuu u
v
w
only log
max 6.9e+10E : 1.e+11D : 2.e+10C : 1.e+10B : 2.e+9A : 1.e+9z : 2.e+8y : 1.e+8x : 2.e+7w : 1.e+7v : 2.e+6u : 1.e+6t : 200000.s : 100000.r : 20000.q : 10000.p : 2000.o : 1000.n : 200.m : 100.l : 20.0k : 10.0j : 2.00i : 1.00h : 0.20g : 0.1f : 0.02e : 0.01
d : 0.002c : 0.001b : 2.e-4a : 1.e-4min 6.94e-5
(a) Electric field distribution
with cpr1: Cycle=1 Time= 50.000 dt= 50.000 p2 Nodes=15863 Cells=7656 RMS Err= 0.0323
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a a
aa
a
a
aa
b
b
c
c
d
d
e
e
f
f
g
g
h
h
i
i
j
j
k
kl m
nop q r
st
u v
v
max 6.56v : 6.30u : 6.00t : 5.70s : 5.40r : 5.10q : 4.80p : 4.50o : 4.20n : 3.90m : 3.60l : 3.30k : 3.00j : 2.70i : 2.40h : 2.10g : 1.80f : 1.50e : 1.20d : 0.90c : 0.60b : 0.30a : 0.00min 0.00
Scale = E3
(b) Potential distribution lines
Fig. 5.2 The electric field distribution (log scale) and potential distribution lines at 50 ns after initial discharge (with Cpr)
Ch.5 The electric field of API plasma electrode system 59
with cpr1: Cycle=3 Time= 150.00 dt= 50.000 p2 Nodes=64193 Cells=31384 RMS Err= 8.2e-5
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
a
a
a
aa
m
m
mm
mn
nn
o
p
p
p
p
p
q
qq q
q
q
q
q q
q
q q
q
r
r
r
r
r
r
r
s
s
s
ss t
t
t
t
t
t
tt
u
u
only log
max 3.3e+11E : 1.e+12D : 2.e+11C : 1.e+11B : 2.e+10A : 1.e+10z : 2.e+9y : 1.e+9x : 2.e+8w : 1.e+8v : 2.e+7u : 1.e+7t : 2.e+6s : 1.e+6r : 200000.q : 100000.p : 20000.o : 10000.n : 2000.m : 1000.l : 200.k : 100.j : 20.0i : 10.0h : 2.00g : 1.00f : 0.20e : 0.1
d : 0.02c : 0.01b : 0.002a : 0.001min 3.32e-4
(a) Electric field distribution
with cpr1: Cycle=3 Time= 150.00 dt= 50.000 p2 Nodes=64193 Cells=31384 RMS Err= 8.2e-5
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
aa a
a a a
a
a
aa
a
a
a
b
c
c
d
d
e
e
f
fg
g
h
h i
i
jklm
n o
pq
v
max 1.63q : 1.60p : 1.50o : 1.40n : 1.30m : 1.20l : 1.10k : 1.00j : 0.90i : 0.80h : 0.70g : 0.60f : 0.50e : 0.40d : 0.30c : 0.20b : 0.10a : 0.00min 0.00
Scale = E4
(b) Potential distribution lines
Fig. 5.3 The electric field distribution (log scale) and potential distribution lines at 150 ns after initial discharge (with Cpr)
Ch.5 The electric field of API plasma electrode system 60
with cpr1: Cycle=7 Time= 350.00 dt= 50.000 p2 Nodes=64211 Cells=31392 RMS Err= 5.2e-6
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
a
a a
m
mm
m
m
n
n
n
nn
n
o
o o
o
o
o
p
p
pp
q q q
q
q
q q
q q
q
q
q r
r
r r
r
r
r
s
s
s
s
s s
t t
t t
t
u
v
only log
max 2.5e+10E : 1.e+11D : 2.e+10C : 1.e+10B : 2.e+9A : 1.e+9z : 2.e+8y : 1.e+8x : 2.e+7w : 1.e+7v : 2.e+6u : 1.e+6t : 200000.s : 100000.r : 20000.q : 10000.p : 2000.o : 1000.n : 200.m : 100.l : 20.0k : 10.0j : 2.00i : 1.00h : 0.20g : 0.1f : 0.02e : 0.01
d : 0.002c : 0.001b : 2.e-4a : 1.e-4min 2.49e-5
(a) Electric field distribution
with cpr1: Cycle=7 Time= 350.00 dt= 50.000 p2 Nodes=64211 Cells=31392 RMS Err= 5.2e-6
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
aa a
a
aaa
a
b
c
d
e
f g
hijk lmn
n
o
o
p
pq qr
r
s
st
u
uv
w
x
x
y
y
zz
z z
z
v
max 0.00z : 0.00y : -0.05x : -0.10w : -0.15v : -0.20u : -0.25t : -0.30s : -0.35r : -0.40q : -0.45p : -0.50o : -0.55n : -0.60m : -0.65l : -0.70k : -0.75j : -0.80i : -0.85h : -0.90g : -0.95f : -1.00e : -1.05d : -1.10c : -1.15b : -1.20a : -1.25min -1.25
Scale = E3
(b) Potential distribution lines
Fig. 5.4 The electric field distribution (log scale) and potential distribution lines at 350 ns after initial discharge (with Cpr)
Ch.5 The electric field of API plasma electrode system 61
without cpr3: Cycle=1 Time= 50.000 dt= 50.000 p2 Nodes=22351 Cells=10888 RMS Err= 0.0222
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
aa
a
a a
a
cd
d
d
d
d
dd
d
e e
e
e
f
f
f
f
g
g
g
g
g
gg
g g
gh
h
h
h
h
h
h
hh
hh
h
i
i
ij
j
jj
jj
j
jk k
k
k
only log
max 1.1e+11u : 2.e+11t : 1.e+11s : 2.e+10r : 1.e+10q : 2.e+9p : 1.e+9o : 2.e+8n : 1.e+8m : 2.e+7l : 1.e+7k : 2.e+6j : 1.e+6i : 200000.h : 100000.g : 20000.f : 10000.e : 2000.d : 1000.c : 200.b : 100.a : 20.0min 11.3
(a) Electric field distribution
without cpr3: Cycle=1 Time= 50.000 dt= 50.000 p2 Nodes=22351 Cells=10888 RMS Err= 0.0222
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
bcd e
fg
h
ij
k ln
n
n
n o
p
p
q
q
r
rs
s
t
t
u
u
v
v
w
w
x
x y
zA
A
v
max 6.60A : 6.50z : 6.00y : 5.50x : 5.00w : 4.50v : 4.00u : 3.50t : 3.00s : 2.50r : 2.00q : 1.50p : 1.00o : 0.50n : 0.00m : -0.50l : -1.00k : -1.50j : -2.00i : -2.50h : -3.00g : -3.50f : -4.00e : -4.50d : -5.00c : -5.50b : -6.00a : -6.50
min -6.59
Scale = E3
(b) Potential distribution lines
Fig. 5.5 The electric field distribution (log scale) and potential distribution lines at 50 ns after initial discharge (without Cpr)
Ch.5 The electric field of API plasma electrode system 62
without cpr3: Cycle=3 Time= 150.00 dt= 50.000 p2 Nodes=95573 Cells=47092 RMS Err= 7.5e-5
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
aaa
a
a
a
a
aa
a
a
b
b
b
c
c
c
c
dd
d
e
e
ee
f
f
f
g gg
g g
g
g
gg
g
g
g
h h
h
h h
h
h
ii
ii
i
i i
j
j
j
k k
only log
max 6.6e+11u : 1.e+12t : 2.e+11s : 1.e+11r : 2.e+10q : 1.e+10p : 2.e+9o : 1.e+9n : 2.e+8m : 1.e+8l : 2.e+7k : 1.e+7j : 2.e+6i : 1.e+6h : 200000.g : 100000.f : 20000.e : 10000.d : 2000.c : 1000.b : 200.a : 100.min 44.6
(a) Electric field distribution
without cpr3: Cycle=3 Time= 150.00 dt= 50.000 p2 Nodes=95573 Cells=47092 RMS Err= 7.5e-5
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
bc
d
e
fg
h
i
ii
i
i ii
i
i
j
j
k
k l
l
m
m
n
n
op q
q
v
max 1.63q : 1.60p : 1.40o : 1.20n : 1.00m : 0.80l : 0.60k : 0.40j : 0.20i : 0.00h : -0.20g : -0.40f : -0.60e : -0.80d : -1.00c : -1.20b : -1.40a : -1.60min -1.63
Scale = E4
(b) Potential distribution lines
Fig. 5.6 The electric field distribution (log scale) and potential distribution lines at 150 ns after initial discharge (without Cpr)
Ch.5 The electric field of API plasma electrode system 63
without cpr3: Cycle=7 Time= 350.00 dt= 50.000 p2 Nodes=95573 Cells=47092 RMS Err= 5.3e-6
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
a
a
a
a
a
a
aaaa
a
c
c
c
d
ddd
d
ee
e
e
e
ee
f
f
f
f
ff
ggg
g g
g
gg
g
gg
gg
h
hh h
h
h
h
i
i
i
ii
i
j
j
k k
only log
max 5.e+10u : 1.e+11t : 2.e+10s : 1.e+10r : 2.e+9q : 1.e+9p : 2.e+8o : 1.e+8n : 2.e+7m : 1.e+7l : 2.e+6k : 1.e+6j : 200000.i : 100000.h : 20000.g : 10000.f : 2000.e : 1000.d : 200.c : 100.b : 20.0a : 10.0min 2.95
(a) Electric field distribution
without cpr3: Cycle=7 Time= 350.00 dt= 50.000 p2 Nodes=95573 Cells=47092 RMS Err= 5.3e-6
X e-2
-3. -2. -1. 0. 1. 2. 3.
Y
e-2
-3.
-2.
-1.
0.
1.
2.
3.
a
a
b
b
c
c
d
d
e
e
f
f
g
g
h
h
ij
j
k
k
l
l
m
m
m
m
nsu
xy
v
max 1.26y : 1.20x : 1.10w : 1.00v : 0.90u : 0.80t : 0.70s : 0.60r : 0.50q : 0.40p : 0.30o : 0.20n : 0.10m : 0.00l : -0.10k : -0.20j : -0.30i : -0.40h : -0.50g : -0.60f : -0.70e : -0.80d : -0.90c : -1.00b : -1.10a : -1.20min -1.25
Scale = E3
(b) Potential distribution lines
Fig. 5.7 The electric field distribution (log scale) and potential distribution lines at 350 ns after initial discharge (without Cpr)
Ch.5 The electric field of API plasma electrode system 64
To compare the previous two experimental results (Fig. 4.6: 5 atm, with Cpr;
and Fig. 4.9: 5 atm, without Cpr) we evaluate the simulation results (Fig. 5.3(a) and
Fig. 5.6(a)). In the simulation results, the main discharge forms around the electric
field region given by the distribution line t for the electrode with Cpr (Fig. 5.3(a)) and
around the line j for the electrode without Cpr (Fig. 5.6(a)). Quantitatively, the
electric field intensities of the distribution line t in Fig. 5.3(a) and line j in Fig. 5.6(a)
the same, representing values above 2 × 106 V/m. The area enclosed by the electric
field distribution line j in Fig. 5.6(a) for the electrode without Cpr is more extensive
than that enclosed by the line t in Fig. 5.3(a) for the electrode with Cpr. This result
seems to indicate an increase in the discharge region (see Fig 5.8). Therefore, in the
case for the electrode without Cpr, it can be assumed that the probability of cathode
spots forming at the edge of metal electrode is increased. For this reason, the main-
discharge volume in Fig. 4.9 deflected to the right side and the maximum energy
deposition was limited in the API plasma electrode discharge system without Cpr.
However, for the electrode with Cpr, a uniform main-discharge was formed by
controlling the effect of capacitance Cpr.
Ch.5 The electric field of API plasma electrode system 65
(a) With Cpr
(b) Without Cpr
Fig. 5.8 Compare the electric field distribution (log scale) with the experimental result at 150 ns after initial discharge
Ch.5 The electric field of API plasma electrode system 66
Reference
[1] FlexPDETM, Finite element software, PDESolutions Inc., Web page: http://www.pdesolutions.com
Ch.6 Characteristics of discharge pumped Argon gas excitation 67
6. Characteristics of discharge pumped Argon gas excitation
Most of the reactions due to electron-atom, ion-electron collisions and the
formations of ion or excited states etc. in high-pressure argon plasma have been
reported (e.g. [1-14]). The principal kinetic processes between electrons and heavy
particles, and the key reactions among the heavy particles are tabulated in tables 6.1
and 6.2 respectively. The important absorption, spontaneous emission and radiation
processes are listed in table 6.3. All of the rate constants and cross sections are
obtained from literature [1-14]. Generally speaking, argon atoms, such as those in
discharge Xe2* dimer kinetics [14], are ionized by multi-step processes to provide
the electron source. The energy of the electrons is acquired from the applied electric
field for the discharged pumped case. When an e-beam is used for pumping, the
argon atoms can be also directly ionized by energetic electrons of the e-beam. The
electrons decay through two-body or three-body electron-ion recombination
processes (see reactions 18 and 19 in table 6.1). The dimers are formed by three-
body collisions of Ar* with 2Ar (see reactions 5 and 6 in table 6.2). In this way, the
kinetic scheme is simplified considerably, but it is still sufficient for a detailed
description of the discharge characteristics [2]. Cross sections of these two effective
electron levels are equal to the sum of the cross sections with similar thresholds.
Ch.6 Characteristics of discharge pumped Argon gas excitation 68
Table 6.1 Electron and Ar gas reactions.
I 1 2 3 4 5 6 7 8 9
10 11 12 13 14 15 16 17
II 18 19
III 20 21 22 23 24 25 26 27
Electron collisions Ar + e → Ar* + e Ar + e → Ar** + e Ar + e → Ar+ + e + e Ar* + e → Ar** + e Ar** + e → Ar+ + e + e Ar* + e → Ar+ + e + e Ar2** + e → Ar2
+ + e + e Ar2*(3Σu) + e → Ar2
+ + e + e Ar2*(1Σu) + e → Ar2
+ + e + e Ar2*(1Σu) + e → Ar + Ar + e Ar2*(3Σu) + e → Ar + Ar + e Ar2*(1Σu) + e → Ar* + Ar + e Ar2*(3Σu) + e → Ar* + Ar + e Ar2*(1Σu) + e → Ar2** + e Ar2*(3Σu) + e → Ar2** + e Ar2*(3Σu) + e → Ar2*(1Σu) + e Ar2
+ + e → Ar+ + Ar + e
Electron-ion recombination Ar3
+ + e → Ar** + Ar + Ar Ar2
+ + e → Ar** + Ar
Electron ion formation and destruction Ar+ + e → Ar+ (4s) + e Ar+ + e → Ar+ (4p) + e Ar+ + e → Ar+ (4d) + e Ar+ (4s)+ e → Ar+ (4p) + e Ar3
+ + e → Ar** + Ar + Ar + e Ar + e → Ar+ (4s) + e + e Ar + e → Ar+ (4p) + e + e Ar + e → Ar+ (4d) + e + e
Ch.6 Characteristics of discharge pumped Argon gas excitation 69
Table 6.2 Key reactions among heavy species
Reactions Forward rate constant cm3s-1 or cm6 s-1
I 1 2
II 3 4 5 6
7 8
III 9
10 IV
11 12 13 14 15 16 17 18 19 20 21
V 22 23 24
Ion formation Ar+ + 2Ar → Ar2
+ + Ar Ar2
+ + 2Ar → Ar3* + Ar , F(T)=(300/Tg)1.5 Excited stats kinetics
Ar** + 2Ar →Ar2** + Ar Ar2** + Ar → Ar* + 2Ar Ar* + 2Ar → Ar2*(3Σu) + Ar Ar* + 2Ar → Ar2*(1Σu) + Ar
Ar2*(3Σu) + Ar → Ar2*(1Σu) + Ar
Ar** + Ar → Ar* + Ar Vibration-Translational relaxation Ar2*(3Σu) + Ar → Ar2*(3Σu)v=0 + Ar Ar2*(1Σu) + Ar → Ar2*(1Σu)v=0 + Ar Penning ionization Ar2*(1Σu) + Ar2*(1Σu) → Ar2
+ + 2Ar + e Ar2*(1Σu) + Ar2*(3Σu) → Ar2
+ + 2Ar + e Ar2*(3Σu) + Ar2*(3Σu) → Ar2
+ + 2Ar + e Ar* + Ar* → Ar+ + Ar + e Ar** + Ar** → Ar+ + Ar + e Ar* + Ar* → Ar2
+ + 2Ar + e Ar2*(1Σu) + Ar* → Ar2
+ + 2Ar + e Ar2*(3Σu) + Ar* → Ar2
+ + 2Ar + e Ar* + Ar** → Ar+ + Ar + e Ar* + Ar2* → Ar+ + Ar + e Ar** + Ar2** → Ar+ + Ar + e Excited ion formation and destruction Ar+(4s) + 2Ar → Ar2
+ + Ar Ar+(4p) + 2Ar → Ar2
+ + Ar Ar+(4d) + 2Ar → Ar2
+ + Ar
2.5 × 10-31(300/Tg)1.5
7.0 × 10-31(300/Tg)1.5
2.5 × 10-32
1.0 × 10-11
1.0 × 10-32/f(Ne/NAr)
3.0 × 10-34/f(Ne/NAr) with f(Ne/NAr)=1+Yr×(Ne/NAr)
and Yr=6×104, Ref [15] 4.0 × 10-14
5.0 × 10-12
9.0 × 10-11
9.0 × 10-11
5.0 × 10-10
5.0 × 10-10
5.0 × 10-10
5.0 × 10-10
5.0 × 10-10
5.0 × 10-10
6.0 × 10-10
6.0 × 10-10
5.0 × 10-10
5.0 × 10-10
5.0 × 10-10
2.5 × 10-31
2.5 × 10-31
2.5 × 10-31
Ch.6 Characteristics of discharge pumped Argon gas excitation 70
Table 6.3 Absorption, spontaneous emission and laser radiation. Photon1-4 are photons released when ionic Ar in various states transits to a lower state. Laser radiation at 126nm is represented as rad in reaction 4 Reactions Rate constant or cross section
1 Ar2*(1Σu) → 2Ar + hv (λ = 126 nm) 2 Ar2*(3Σu) → 2Ar + hv (λ = 126 nm) 3 Ar** → Ar* + hv1 4 Ar2*(1Σu) + rad → Ar + Ar + rad
5 Ar* → Ar + hv1
6 Ar2*(3Σu) + hv (λ = 126 nm) → Ar2+ + e
7 Ar+(4s) → Ar+ + Photon1 8 Ar+(4p) → Ar+ (4s) + Photon2
9 Ar+(4d) → Ar+ (4p) + Photon3 10 Ar+(4d) → Ar+ + Photon4
2.38 × 108 Lifetime 4.2ns 3.13 × 105 Lifetime 3.2µs
1.40 × 107
1.24 × 10-17/W(Tg) with W(Tg) = 1+ 0.27 (Tg/300) exp[-0.24(Tg/300)2] 5.00 × 104
5.00 × 10-19
2.64 × 109
1.19 × 108
2.30 × 108
3.60 × 109
6.1 Light emission characteristics of discharge pumped Ar gas
In order to clarify the light emission characteristic of the API discharge system of
Fig. 4.4 with uniform main-discharge in Ar gas, the light emission with the
wavelength 427.8 nm and 696.5 nm are measured [1]. To measure the emission
characteristic from the Ar gas discharge, a grating monochromator (CT-25C, JASCO)
and a photomultiplier tube (R372, Hamamatsu, Japan) were used. A He-Ne laser as
shown in Fig. 6.1 was used for the arrangement of the electrode and the calibration of
a monochromator. The gas used for these measurements was pure Ar gas. The
experiments were performed at a repetition rate of 1 Hz.
Ch.6 Characteristics of discharge pumped Argon gas excitation 71
CathodeAnode
CHAMBER
Power Circuit
Powercontroller
PMT
Oscilloscope
MONOCHROMATOR
He - Ne Laser
H.V. probe
Mirror
Fig 6.1 Schematic for the observation of the light emission characteristics in Ar gas discharge
Figure 6.2 shows the diagram of energy levels of major transitions in Ar gas and
the reaction processes of Ar2* excimer band (around 126 nm). Here, the wavelength
427.8 nm is Ar ionic line and the wavelength 696.5 nm is an Ar atomic line. The
kinetic reactions between the radiation processes of these two wavelengths and the
radiation processes of Ar2* excimer band (around 126 nm) can be considered simply
as tables 6.1 ~ 6.3. It might be inferred from radiation deexcitation reactions
Ar2*(3Σu) → hv(VUV) + 2Ar (Lifetime 3.2µs) and Ar2*(1Σu) → hv(VUV) + Ar
(Lifetime 4.2ns), mixing process reactions Ar2*(3Σu) + e ↔ Ar2*(1Σu) + e and
Ar2*(1,3Σu) ↔ Ar2** + e that Ar2* (around 126 nm) emission intensity as a function
of time has more than at least two peaks in the case of discharge-pumped schemes.
Ch.6 Characteristics of discharge pumped Argon gas excitation 72
Ar 1s0
Ar+
Ar**
Ar*
ARGON ATOM
ARGON MOLECULE
+e
16
14
12
10
0
Ar2+
Ar2** + e
+e
+ 2Ar
+ e
+ e
+ e
+ 2Ar
+ e , Ar
+ e , Ar*
+ 2Ar Ar2*
+ e
ENER
GY[
eV]
+ e
+ e
+ e
45
37
34
+ e
Pumping
+e +e +e
Ar3+
427.8 nm
696.5 nm
126 nm
Ar++
Collisional relaxation or excitation with the particles indicated
Spontaneous dissociation
Radiative deexcitation
sec13.020.41 nu ±∑ +
sec3.02.33 µ±∑ +u
∑ +g
1
Fig 6.2 The diagram of energy levels of major transitions in Ar gas
Ch.6 Characteristics of discharge pumped Argon gas excitation 73
Such experimental results are demonstrated in the literatures [1,15]. It is likely
that only the radiation deexcitation reaction Ar2*(1Σu) → hv(VUV) + Ar, leads to
Ar2* excimer laser action. However, the rate constant of Ar2*(3Σu) is similar to that
of Ar2*(1Σu) [16], the radiative lifetime of Ar2*(3Σu) is 3.2 µs which is much longer
than the case of Ar2*(1Σu). Therefore the energy level of Ar2*(3Σu) is easy to become
saturated. In addition, this energy level reacts with VUV (vacuum ultraviolet) light as
the photo-ionization reaction Ar2*(3Σu) + hv (VUV) → Ar2+ + e. Excimer formation
reactions Ar* + 2Ar → Ar2*(3Σu) + Ar and Ar* + 2Ar → Ar2*(1Σu) + Ar are most
important processes for the realize Ar2* excimer laser action. It is important to study
the generation process of excited Ar* (Ar atom of first excited level is most
important). The excited Ar* atoms were formed by the lower excited energy level
about 11.6 eV. i.e. the reactions Ar + e → Ar*(4s) + e, Ar2+ + e → Ar* +Ar, Ar2* + e
→ Ar* + Ar + e, Ar2** + Ar → Ar* + 2Ar and Ar3+ + e → Ar* + 2Ar. However,
excited Ar* atoms which were formed by the reactions Ar2+ + e → Ar* +Ar, Ar2* + e
→ Ar* + Ar + e, Ar2** + Ar → Ar* + 2Ar and Ar3+ + e → Ar* + 2Ar are not
suitable for the formation and enhancement of lasering band Ar2*(1Σu) because these
Ar* atoms were formed late comparatively. Therefore, for the formation and
enhancement of lasering band Ar2*(1Σu), it is suitable that it is increased the electron
impact excitation Ar + e → Ar*(4s) + e by very fast pumping energy deposition in
the uniform main-discharge condition.
Figure 6.3 illustrates the variation of the emission intensities of the Ar ionic line
at 427.8 nm and Ar atomic line at 696.9 nm with Ar gas pressure obtained from the
main-discharge. The obtained result is quite similar to the experimental results of the
Ch.6 Characteristics of discharge pumped Argon gas excitation 74
literature [1]. According to the literature, the waveform of the ionic line at 427.8 nm
was observed in the discharge and the waveform of the atomic line at 696.5 nm was
observed both in the discharge and in the decaying plasma.
Ch.6 Characteristics of discharge pumped Argon gas excitation 75
0 1 2 3 4 5 6 7 8 9 10Time[us]
Dis
char
ge v
olta
ge [K
V]
696.5nm
0
0
0IN
TEN
SIT
Y(ar
b. u
nits
)IN
TEN
SIT
Y(ar
b. u
nits
)
4atm5atm
2atm 3atm
1atm
-20
-10
4atm5atm
2atm3atm
1atm
427.8nm
1atm2atm
3atm
4atm
5atm
1.8 1.9 2.0 2.1 2.2 2.3 2.4Time[us]
1 2 3 4 5 6 7 8 9 10Time[us]
Fig. 6.3 Variation of the emission intensities of the Ar ionic line at 427.8 nm and Ar atomic line at 696.9 nm in Ar gas (experiment of Fig. 4.3)
Ch.6 Characteristics of discharge pumped Argon gas excitation 76
These results are in agreement with the experimental results of Fig. 6.3. That is,
in the case of 1 atm, the waveform of the ionic line at 427.8 nm has the FWHM (full
width at half maximum) of about 700 ns and an effective lifetime of about 3 µs, and
the waveform of the ionic line at 696.5 nm has the FWHM of about 850 ns and an
effective lifetime of about 6 µs. It can be considered that the waveform of the ionic
line at 696.5 nm shows the formation of the excited Ar* atom from higher excited
state atoms or ions, by a radiative transition or a recombination process. Figure 6.4
shows the dependence of the light emission intensities of the Ar ionic line at 427.8
nm and Ar atomic line at 696.9 nm with the Ar gas pressure. The intensities of the
line at 427.8 nm and the line at 696.9 nm increase with increasing Ar gas pressure.
The intensity of the line at 427.8 nm increases proportionally to the approximate
cubic of the Ar gas pressure, i.e. a curve fitting Y = 0.22X3, where Y is emission
intensity and X is the Ar gas pressure (atm). On the other hand, the line intensity at
696.9 nm shows saturation at pressure above 4 atm. It can be considered that Ar
particles are pumped uniformly by the uniform main-discharge. In addition, it is
considered that the electron collision processes is enhanced due to the increase of the
pumping energy in the uniform main-discharge as shown in Fig. 4.8 which is
obtained experimentally.
Ch.6 Characteristics of discharge pumped Argon gas excitation 77
0
10
20
30
1 2 3 4 5Ar gas pressure [atm]
INTE
NSI
TY [a
rb.u
nits
]
λ=427.8nm λ=696.5nm
Fig. 6.4 Dependence of the emission intensities of the Ar ionic line at 427.8 nm and Ar atomic line at 696.9 nm with the Ar gas pressure, which is derived from Fig. 6.14. Solid line denotes the fitting curve of Y =0.22X3, where Y is intensity and X is the Ar gas pressure (atm)
Consequently, these experimental results suggest that the lasering band
Ar2*(1Σu) is enhanced due to the increase of the excited Ar* atom which is formed by
the electron collision reaction Ar + e→Ar* + e and the Ar2*(3Σu) is saturated by the
recombination process.
Ch.6 Characteristics of discharge pumped Argon gas excitation 78
Reference [1] H. Ninomiya and K. Nakamura, Ar2* excimer emission from a pulsed electric discharge in pure Ar gas, Opt. Commun. 134, 521 (1997) [2] K. S. Gochelashvily, A. V. Demyanov, I. V. Kochetov and L. R. Yangurazova, Fluorescence model of noble gas dimers in pulsed self-sustained discharges, Opt. Commun. 91, 66 (1992) [3] S Neeser, T Kunz and H Langhoff, A kinetic model for the formation of Ar2 excimers, J. Phys. D: Appl. Phys. 30, 1489 (1997) [4] H.A. Koehler, L.J. Ferderber, D.L. Redhead, and P.J. Ebert, Stimulated VUV emission in high-pressure xenon excited by high-current relativistic electron beams, Appl. Phys. Lett. 21, 198 (1972) [5] E Elson and M Rokni, An investigation of the secondary electron kinetics and energy distribution in electron-beam-irradiated argon, J. Phys. D: Appl. Phys. 29, 716 (1996) [6] Takefumi Oka, Masuhiro Kogoma, Masashi Imamura, and Shigeyoshi Arai, Energy transfer of argon excited diatomic molecules, J. Chem. Phys. 70, 3384 (1979) [7] J. W. Keto and Chien-Yu Kuo, Cascade production of Ar(3p54p) following electron bombardment, J. Chem. Phys. 74, 6188 (1981) [8] D. L. Turner and D. C. Conway, Study of the 2Ar+Ar2
+=Ar+Ar3+ reaction, J.
Chem. Phys. 71, 1899 (1979) [9] R. S. F. Chang and D. W. Setser, Radiative lifetimes and two-body deactivation rate constants for Ar(3p5, 4p) and Ar(3p5,4p) states, J. Chem. Phys. 69, 3885 (1978) [10] R. Sauerbrey, The photoionization cross sections of the Rg*2[3Σu
+] excimer
Ch.6 Characteristics of discharge pumped Argon gas excitation 79
states for Ne2*, Ar2*, and Kr2*, IEEE J. Quantum Electron. 23, 5 (1987) [11] P. Dubé, M. J. Kiik, and B. P. Stoicheff, Spectroscopic study of vibrational relaxation and cooling of rare-gas excimers formed in a direct current discharge with supersonic expansion, J. Chem. Phys. 103, 7708 (1995)
[12] Wei-cheng F. Liu and D. C. Conway, Ion–molecule reactions in Ar at 296, 195, and 77 °K, J. Chem. Phys. 62, 3070 (1975)
[13] J. M. Hammer and C. P. Wen, Measurements of Electron Impact Excitation Cross Sections of Laser States of Argon(II), J. Chem. Phys. 46, 1225 (1967)
[14] I. V. Kochetov and Dennis Lo, Kinetics of a self-sustained discharge-pumped Xe*2 laser at 172 nm, Opt. Commun. 113, 541 (1995) [15] K. Nakamura, Y. Ooguchi, N. Umegaki, T. Goto, T. Jisuno, T. Kitamura, M. Takasaki, and S. Horiquchi (private communication) [16] W. Sasaki. Rev. Laser. Eng. 13, 912 (1985) (in Japanese)
Appendix #A 80
7. Conclusions
A new pre-ionization electrode was designed using the surface-corona pre-
ionization method. It was investigated experimentally in terms of generated charge
density, electrical characteristics. These experimental results suggest that a sharp
edge of the ground electrode is possible to radiate the strongest UV light. Therefore,
the API plasma electrode discharge system has been designed. The uniform main-
discharge was formed and its volume and the breakdown voltage Vbr increased with
increasing Ar gas pressure. The instantaneous maximum discharge electric power
was 90 MW at 5 atm Ar gas and the maximum energy deposition was 1.4 MW/cm3.
It was demonstrated that the uniform pre-ionization formed the uniform main-
discharge by the control effect of Cpr and the maximum energy deposition was
increased. It was examined the time dependent main-discharge from two-
dimensional simulation of electric field distribution of the API plasma electrode
discharge system. In the case of the using plate electrode on an anode, this electrode
structure is not suitable for the formation of the uniform main-discharge due to the
arc discharge with Ar gas.
It was measured light emission intensity of the Ar ionic line at 427.8 nm and Ar
atomic line at 696.9 nm with the Ar gas pressure. The intensity of the line at 427.8
nm increases proportionally to the approximate cubic of the Ar gas pressure (Y =
0.22X3, where Y is intensity and X is the Ar gas pressure). On the other hand, the line
intensity at 696.9 nm shows saturation at pressure above 4 atm. These experimental
results suggest that the lasering band Ar2*(1Σu) is enhanced due to the increase of the
Appendix #A 81
excited Ar* atom which is formed by the electron collision reaction Ar + e→Ar* + e
and the Ar2*(3Σu) is saturated by the recombination process.
Although these results are not enough to the discharge-pumped Ar2* excimer
laser action, it seems quite probable if a new chamber designed (i.e. it is more
compact and is operated more high pressure than 15 atm). Therefore, we will make a
new chambera and also extend our research for different gas lasers such as Xe2*,
Kr2* or F2.
a It was designed in Appendix #B.
Appendix #A 82
Appendix #A Necessary conditions for the discharge-pumped Ar2* laser
Necessary conditions for the discharge-pumped Ar2* laser, first, the necessary E/P is
given by the simple expressions: According to the gas law (at 0°C , 1atm) V0 = 2.241ⅹ10-2 m3/mole , N = NA/Vo = 2.687ⅹ1019 /cm3 atm , where, NA: Avogadro constant. Electron mean free path λe = (σesN)-1 = 4.652ⅹ10-3 cm atm , here, Ar2
* stimulated emission cross section σes: 8ⅹ10-18 cm2 [Ref. 9]. E/P ≈ (16)V/λe = 3439.38 V/cm atm , where, 16 eV is the energy of a Ar atom ionization
Second, the necessary pumping power for the discharge-pumped Ar2
* laser is given by the below expressions: Ar2
* saturation intensity Isat
,/46102.4108
1037.2106256.6/ 29218
11534
CmMWsCm
ssJhvI useulSAT =×××××⋅×
== −−
−−
τσ
where, 1159
8
1037.210126
/1099.2/ −− ×=
××
== sm
smcv ulul λ ,
c : the speed of light λul : the wavelength of Ar2
* emission h : Planck constant
512 ±≅⋅⋅∆⋅ LkNseσ
where, L: laser medium length ∆N: population density k: coefficient with optical gain
thus,
Necessary pumping power =)(
512CmkL
ISAT±
×
Appendix #A 83
Fig. a.1 Necessary condition for discharge-pumped Ar2* laser with breakdown
voltage Vbr
0
50
100
150
200
250
300
350
400
30 50 70 90 110 130 150 170 190
Laser gain medium length L [cm]
Pum
ping
pow
er [M
W/c
m3 ]
0
50
100
150
200
250
300
350
400
30 50 70 90 110 130 150 170 190
Laser gain medium length L [cm]
Pum
ping
pow
er [M
W/c
m3 ]
Fig. b.1 Necessary pumping power for discharge-pumped Ar2* laser with laser gain
medium length L (k=8%)
Appendix #B 84
Appendix #B Design of a new chamber and electrode
FRONT Unit [mm]
Appendix #B 85
ELECTRODE FRONT
Appendix #B 86
TOP
Appendix #B 87
SIDE
Appendix #B 88
ELECTRODE SIDE
Appendix #B 89
WINDOW
Appendix #B 90
Stre
ss n
umer
ical
cal
cula
tion
Stre
ss n
umer
ical
cal
cula
tion
Appendix #B 91
Stre
ss n
umer
ical
cal
cula
tion
List of publications 92
List of publications
1. S. K. Hong, N. Hayashi, S. Ihara, S. Satoh and C. Yamabe “Formation properties of the main-discharge in pure Ar gas using the automatically pre-ionized plasma electrode” The IEEE Transactions on Plasma Science, Vol.33. Apr (2005). pp.324-325.
2. S. K. Hong, N. Hayashi, S. Ihara, S. Satoh and C. Yamabe “Main-discharge formation and light emission in pure Ar gas at multi-atmospheric pressure using the automatically pre-ionized plasma electrode” Vacuum, Vol.79. Jun (2005), pp. 25-36
3. S. K. Hong, N. Hayashi, S. Ihara, S. Satoh, C. Yamabe and S. B. Wee “The discharge electrode for Ar2* excimer laser using plasma cathode” Optics Communications (in press)
4. J. Morida, S.K. Hong, N. Hayashi, S. Ihara, S. Satoh and C. Yamabe “真空紫外エキシマレーザ励紀用放電システムの基礎特性” Journal of Applied Plasma Science, Vol.11. Dec (2003). pp.129-134 (in Japanese)