UC San Diego
UC San Diego Previously Published Works
Title
Energy absorption and propagation in laser-created sparks
Permalink
https://escholarship.org/uc/item/339818gc
Journal
Applied Spectroscopy, 58(6)
ISSN
0003-7028
Authors
Bindhu, C V
Harilal, S S
Tillack, M S
et al.
Publication Date
2004-06-01
Peer reviewed
eScholarship.org Powered by the California Digital Library
University of California
Volume 58, Number 6, 2004 APPLIED SPECTROSCOPY 719
0003-7028/04/5806-719$2.00/0
q
2004 Society for Applied Spectroscopy
Energy Absorption and Propagation in Laser-Created
Sparks
C. V. BINDHU, S. S. HARILAL,* M. S. TILLACK, F. NAJMABADI, and
A. C. GAERIS
Center for Energy Research, University of California San Diego, 9500 Gilman Drive, La Jolla, California 92093-0438
The energy absorption and laser propagation characteristics of air
and argon sparks at one atmosphere have been investigated. To
create the sparks, 532 nm pulses from a frequency doubled Q-
switched Nd:YAG laser are used. We employed 2 ns gated fast
photography for studying the time evolution of the kernel at early
times. Optical emission spectroscopy is used to infer temperature
and density of the sparks. Significant energy absorption by the plas-
ma is observed just above the breakdown threshold. The energy
absorption and propagation in the spark indicated that argon plas-
ma is more absorptive than air plasma. The absorption of the spark
increases with laser energy, and at higher energies absorption sat-
uration is observed. A spiky behavior is observed in the transmitted
temporal profiles of lasers at higher energies and this is explained
as due to the formation of a self-regulating regime.
Index Headings: Laser-induced gas breakdown; Emission spectros-
copy; Laser propagation; Energy absorption; Spark photography.
INTRODUCTION
When a powerful focused laser beam interacts with a
gaseous medium, a spark is created.
1,2
This high-pressure
region develops a shock wave into the ambient medium
that has sufficient strength to ignite a gaseous mixture
3
or to extinguish a diffusion flame.
4
Laser-induced gaseous
plasmas can be used for a variety of applications includ-
ing elemental analysis,
5,6
detecting airborne biological
agents,
7
quantitative analysis of aerosols,
8
production of
X-rays and soft X-rays,
9,10
ultra fast shutters,
11
etc. Gas
breakdown studies represent the initial step in research in
inertial confinement fusion (ICF) and plasma heating by
laser radiation. Different authors have examined the
shapes and expansion of the laser-induced breakdown
kernels using fast photography,
12
Schlieren photogra-
phy,
13
and laser-induced fluorescence techniques.
14
The
temporal, spatial, and spectral features of a plasma kernel
suggest that it can be used as a pulsed bright and broad-
band ultraviolet-visible light source.
15
The ionization of a gas in the presence of an electro-
magnetic field takes place by different mechanisms such
as multiphoton absorption, cascade ionization via inverse
bremsstrahlung, etc.
16,17
Both of these phenomena are
treated theoretically by different groups.
18,19
In intense
light an electron can gain sufficient energy by absorbing
photons from the radiation field in collision with neutral
atoms to produce ionization. If the laser irradiance is high
enough, the multiphoton ionization process can generate
the initial electrons from which an electron cascade can
develop. Breakdown will occur if the electron density can
Received 23 September 2003; accepted 2 February 2004.
* Author to whom correspondence should be sent. E-mail: harilal@
fusion.ucsd.edu.
reach a critical value despite losses due to diffusion, elec-
tron attachment, etc. A more detailed account of break-
down threshold calculations was given by Kroll and Wat-
son,
18
in which they review existing literature on various
loss processes of importance for breakdown such as dif-
fusion, attachment, and vibrational and electronic exci-
tation. Inverse bremsstrahlung is dominant when the
product of gas pressure and pulse width P
t.
10
2
7
torr
s. For smaller values of P
t
, collisions do not have time
to occur during the laser pulse and multiphoton ionization
is regarded as the dominant mechanism for creating a
spark.
Most of the experiments related to laser-induced gas
breakdown studies have centered largely on the measure-
ments of breakdown threshold, while a little effort has
been made on studies of energy absorption and propa-
gation in a spark.
20–22
Recently, Chen et al.
12
reported la-
ser-induced breakdown and energy deposition in air at
atmospheric pressure using the fundamental frequency
from a Q-switched Nd:YAG laser. Their results give
quantitative guidance for the selection of the laser energy
to achieve the desired breakdown probability and energy
deposition characteristics of the plasma. The present in-
vestigation is designed to study the interaction of a fo-
cused laser pulse with one atmosphere air or argon and
subsequent energy propagation through the spark medi-
um. We analyzed how much energy can be propagated
through a spark medium and how the spark affects the
temporal profiles of the laser. The optical breakdown is
created by focusing 532 nm, 8 ns pulses from a frequency
doubled Q-switched Nd:YAG laser. Fast photography is
undertaken to evaluate the evolution of the plasma kernel
at earlier times. Spectroscopic methods are used for elu-
cidating temperature and density of the sparks.
EXPERIMENTAL SETUP
The experimental arrangement used for the creation of
laser sparks and studies of beam propagation is shown in
Fig. 1. We used pulses from a frequency doubled Q-
switched Nd: YAG laser (full width at half-maximum,
FWHM, 8 ns) for creating breakdown plasma. The laser
pulse parameters were monitored using a photodiode, a
beam profiler (Photon Inc.), and a CLAS 2D wavefront
sensor (Wavefront Sciences). The pulses were temporally
cleaned up by the Quanta Ray injection seeder attach-
ment. The spatial structure of the laser profile was ap-
proximately Gaussian. The pulse energy was varied by
using a combination of wave plate and cube beam splitter.
To create breakdown plasma, the laser pulses were fo-
cused using an f/5 antireflection-coated laser aplanat (CVI
Laser, LAP 75.0–15.0) having a focal length of 75 mm.
720 Volume 58, Number 6, 2004
F
IG
. 1. The schematic of the experimental setup used for imaging, spectroscopy, propagation, and energy absorption studies. (SHG) Second-
harmonic generator; (WP) wave plate; (C) cube beam splitter; (BD) beam dump; (R) reflector; (BS) beam sampler; (F) 532 nm filter; (ICCD)
intensified charge-coupled device; (PTG) programmable timing generator; (L1) high-energy laser aplanat; (L2) lens; (HBS) holographic beam
sampler; (EM) energy meter; (BP) beam profiler; (CLAS2D) CLAS2D Shack Hartman sensor; and (PD) photo diode.
A 75 mm antireflection-coated plano-convex lens was
used for collimating the laser beam after passing through
the focal region. The optical elements were placed in a
vacuum chamber.
Photodiodes (Electro-optics Technology, Model EOT
2000, rise time: 200 ps) and energy/power meters (Ophir)
were used to record temporal profiles and energy of each
incoming and transmitted laser pulse through the focal
volume. The temporal profiles were monitored using a 1
GHz Digital Phosphor Oscilloscope (Tektronix TDS5014,
5 GS/s maximum real-time sample rate).
The kernel imaging was accomplished using an inten-
sified charge-coupled device (ICCD, PI MAX, Model
512 RB) placed orthogonal to the laser propagation di-
rection. A Nikon lens was used to image the plume re-
gion onto the camera to form a two-dimensional image
of the plume intensity. In order to eliminate 532 nm stray
photons reaching the camera, a magenta subtractive filter
was used. A programmable timing generator was used to
control the delay time between the laser pulse and the
imaging system with overall temporal resolution of 1 ns.
For emission spectroscopic studies, an optical system
was used to image the spark onto the entrance slit of a
0.5-m spectrograph (Acton Pro, Spectra-Pro 500i), so as
to have one-to-one correspondence with the sampled area
of the spark and the image. The spectrograph was
equipped with a tri-grating turret (150, 600, and 2400
grooves/mm) and an ICCD camera that was operated
with vertical binning to obtain spectral intensities versus
wavelength.
RESULTS AND DISCUSSION
The breakdown threshold is one of the first parameters
of interest in characterizing the formation of plasma in a
gaseous medium. When a lens focuses the laser beam,
the distribution of irradiance in the focal spot is deter-
mined by the mode structure in the laser oscillator, by
the effect of amplifiers and apertures in the system, and
by the parameters of the lens. Single-element lenses were
used in most of the previous experiments
12,14,20,22–26
for
focusing high-power laser beams and significant spherical
aberration was often present, though it may be minimized
by a suitable choice of curvatures. The presence of ab-
errations may lead to erroneous values of optical break-
down threshold even when the measured spot size is used
for the determination.
27
The effect of spherical aberration
on the distribution of irradiance in the focal region has
been calculated by Evans and Morgan.
28
In order to avoid
or minimize spherical aberration, we used a two-element
laser aplanat for focusing the laser beam. As a rule of
thumb, a lens should be considered perfect or diffraction
limited if the optical path difference (OPD) is less than
¼
wave. The estimated OPD of the laser aplanat used in
the present studies using Zemax image analysis
29
is about
1
⁄
5
wave, so the aberration effects are expected to be in-
significant.
The breakdown threshold is determined by observing
the initiation of the plasma kernel using the ICCD camera
and looking at the absorption in the transmitted laser
pulse through the focal volume. The measured break-
APPLIED SPECTROSCOPY 721
F
IG
. 2. The time evolution of visible emission from an air spark at different laser energies recorded using an ICCD camera. The exposure time
used was 2 ns. The timings in the images represent the time after the onset of plasma formation. All of the images are normalized to their maximum
intensity. The laser is coming from the left-hand side. The time sequence of each row is given along the top and the energy used for creating each
column is given along the left side.
down threshold for air and argon at 1 atmosphere is (2.5
6
0.2)
3
10
12
Wcm
2
2
and (2.3
6
0.2)
3
10
12
Wcm
2
2
,
respectively. The focal spot dimensions (
;
6
m
m) used in
determining the threshold irradiances quoted here are not
measured directly but are calculated from the beam di-
vergence (measured using a CLAS 2D Shack Hartmann
sensor) and the focal spot estimated using Zemax image
analysis. It is difficult to compare breakdown thresholds
directly from the literature as experimental parameters
rarely match. The breakdown threshold value of air is in
good agreement with that previously reported by Phuoc
22
with almost identical experimental conditions.
Spark Imaging. The time evolution of the spark ker-
nel is studied using fast photography and the shapes of
the breakdown kernel observed show interesting behav-
ior. Figures 2 and 3 give the time evolution of air and
argon sparks at 1 atmosphere at different energy levels.
The duration of the intensification (exposure time) is 2
ns and each image in the figure is recorded from an in-
dependent breakdown event. The laser beam is incident
from the left-hand side. It should be remembered that
each image given in the figures is spectrally integrated in
the region 350–900 nm due to emission from the excited
states of various species. They are not necessarily rep-
resentative of the total flux because a part of the plume
is nonluminous.
30,31
For better clarity, all of the images
are normalized to their maximum intensity. This gives a
fast on-scale overview of the data.
The shapes of the air spark (Fig. 2) at early times and
at low energies are spherical and are confined very close
to the focal volume. But as time elapses the spherical
shape is changed to an elliptical shape. With increasing
laser energy, the kernel becomes more asymmetrical in
shape; the backward-moving plasma (towards the focus-
ing lens) grows much faster than the forward-moving
plasma (away from the focusing lens). The subsequent
radiation expansion of the hot plasma leaves a rarefied
region in the focal volume. The initiation time is earlier
and the rate of growth is faster if the pulse energy is
increased. The layer of gas outside the plasma, although
it is transparent to the laser beam, is heated by the plasma
radiation. This outside gas close to the plasma will in
turn be ionized to such an extent that it will strongly
absorb the laser light. This layer will then be further heat-
ed very rapidly and the temperature increases. By this
time a new layer of plasma nearer the laser will have
become strongly absorbing, so the boundary of the plas-
ma will move towards the focusing lens. The absorption
of the laser photons by the plasma is mainly due to the
inverse bremsstrahlung process, which is so dominant
that it leads to the development of a laser-supported ra-
diation wave that propagates toward the laser beam. The
time scale of this event is that of the laser pulse itself.
The time evolution of the laser-created argon spark (Fig.
3) at different energy levels showed remarkable differ-
ences in comparison with the air spark. At low energy
levels (near the breakdown threshold), the behavior of the
argon spark resembled the air spark. But as the energy
of the pulse increases, the forward-moving spark com-
ponent is totally absent. After the initial breakdown, the
plasma is heated to a point where it is opaque to the
incoming laser beam. The optically thick plasma absorbs
practically all the incident radiation after breakdown. We
used the imaging data to create position–time (R–t) plots
and the velocities of the sparks moving towards the fo-
cusing lens are measured from the slopes of the R–t
graph. The estimated velocity of spark propagation to-
wards the laser beam is given in Fig. 4 for air and argon
for different laser energies. From the figure it is evident
that the velocity of the spark front increases with laser
energy.
In the present studies, the incident laser irradiance used
is clearly greater than 1 GWcm
2
2
and for those cases for
which the absorbed energy significantly exceeds the
breakdown threshold, the plasma propagates as a laser-
supported radiation wave. For a laser-supported radiation
wave, the velocity is given by
32
v
5
I/
r
E, where I is the
incident laser intensity,
r
is the mass density of the gas,
and E is the internal energy. As the mass density of the
air at 1 atm (1.205
3
10
2
3
g/mL) is smaller than the
density of argon at 1 atm (1.664
3
10
2
3
g/mL), it is
expected that a laser-supported radiation wave in air will
propagate faster than one in argon. This is also supported
by the fact that the ratio of the velocities of air and argon
722 Volume 58, Number 6, 2004
F
IG
. 3. ICCD photographs of the time evolution of visible emission from a laser-produced argon spark at different laser energies. The experimental
parameters are the same as described in Fig. 2. All of the images are normalized to their maximum intensity.
F
IG
. 4. The estimated velocity of the spark propagation towards the
laser beam with laser energy for argon and air sparks showing that the
air spark is moving faster than the argon spark. The solid curve in the
plot represents the propagation velocity calculated from the scaling law
of the laser-supported radiation wave.
sparks at a particular laser energy matches well with the
ratio of their mass densities. The scaling law for the prop-
agation velocity of the laser-supported radiation wave
is:
32
v
}
I
4/(
b1
4)
(1)
where
b
is a constant whose value varies between 1.5
and 1.6 for most ionized gases. A fit with the above equa-
tion is also given in Fig. 4, which is in good agreement
at higher laser energies.
In air and argon sparks, a cascade-like growth of ion-
ization is evident. The electron cascade requires the ex-
istence of initial electrons. The electrons then absorb
more photons via the inverse bremsstrahlung process. A
cascade growth of electron energy and the absorption co-
efficient of the plasma will be greatly influenced by the
nature of the gas used. The condition necessary for the
cascade-like growth is given by:
32
22
d
«
4
p
eI
n
2m
n
E
eff e eff
52.
0 (2)
2
dtnc
v
M
e
where
«
is the energy of the free electrons; e and m
e
are
the charge and mass of the electron; M is the mass of the
background gas neutral particle; E is the energy of the
first ionization state of the gas;
n
eff
is the effective fre-
quency of electron neutral collision; n
e
is electron den-
sity; I is the radiation intensity; and
v
is the cyclic fre-
quency of radiation. The first term on the right-hand side
represents the rate of growth of energy by the absorption
of laser photons and the second term gives the maximum
rate of energy loss due to elastic and inelastic collisions
with neutral gas particles. In comparing argon and air
atmospheres, cascade growth is more favored for argon
(M
5
40 and E
5
15.75 eV) in comparison with air (for
nitrogen, M
5
14 and E
5
14.54 eV). So the spark
formed in argon is more absorptive than that formed in
air. This is also supported by the fact that the forward
component of the kernel is absent for the argon spark.
The expansion of the spark continues even after the
end of laser pulse, though much more slowly. The time
evolution of the spark at later times showed dissipation
times of
;
5 ms.
33
It was noticed that the shape of the
sparks changes significantly at later times.
33,34
Spectroscopic Studies. For the study of absorption be-
havior of the laser-created sparks, the fundamental spark
parameters are necessary. In order to determine the den-
sity and temperature of the sparks, spectroscopic methods
are used.
17,35,36
The plasma electron temperature was de-
duced from the Boltzmann plot method and for its deter-
mination it was necessary to assume that the sparks pro-
duced are in local thermodynamic equilibrium (LTE).
Several N
1
and Ar
1
lines were selected in order to cal-
culate electron temperatures of air and argon sparks, re-
spectively. The designation and other spectroscopic con-
stants used for determination of the excitation tempera-
ture for air and argon sparks by the Boltzmann plot meth-
od are shown in Table I. Transition probabilities (A
ij
) and
statistical weights (g) of these lines are obtained from the
literature.
37,38
The plasma electron density was measured
using Stark broadened profiles of singly charged ions in
the spark. For the air spark, the stark broadening of N
1
