Critical Reviews in Analytical Chemistry

ISSN: 1040-8347 (Print) 1547-6510 (Online) Journal homepage: https://www.tandfonline.com/loi/batc20

Fundamentals and Applications of Laser-Induced

Breakdown Spectroscopy

D. A. Rusak , B. C. Castle , B. W. Smith & J. D. Winefordner

To cite this article: D. A. Rusak , B. C. Castle , B. W. Smith & J. D. Winefordner (1997)

Fundamentals and Applications of Laser-Induced Breakdown Spectroscopy, Critical Reviews in
Analytical Chemistry, 27:4, 257-290, DOI: 10.1080/10408349708050587

To link to this article: https://doi.org/10.1080/10408349708050587

Published online: 23 Oct 2006.

Submit your article to this journal

Article views: 1957

View related articles

Citing articles: 47 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=batc20
 Critical Reviews in Analytical Chemistty, 27(4):257-290 (1997)

Fundamentals and Applications

of Laser-Induced Breakdown Spectroscopy
D. A. Rusak, B. C. Castle, 6.M! Smith, and J. D. Winefordner*
Department of Chemistry, University of Florida, Gainesville, FL 32611-7200

* Author to whom correspondence should be sent.

ABSTRACT: Laser-induced breakdown spectroscopy is reviewed by dividing the literature into

three categories according to target phase: solid, liquid, or gas. Within each category, the
literature is further divided into a fundamental studies section and an analytical results and
applications section.

KEY WORDS: laser-induced breakdown spectroscopy.

1. INTRODUCTION experiments have, for the most part, been

omitted under the assumption that these pa-
There have been several publications in pers are more pertinent to materials science
the last decade which have reviewed laser- than to analytical chemistry. Fundamentals
induced breakdown spectroscopy (LIBS) and applications of laser ablation as a sam-
either as a unique method of elemental analy- pling technique are also ignored on the
sis, or as a member of the family of atomic grounds that, in these instances, the laser
emission techniques. LIBS has also been plasma does not serve as the excitation
addressed in reviews on laser applications, source.
process monitoring, and materials process- The time frame to be covered has not
ing. Many spectroscopic studies of laser plas- been specifically defined because the amount
mas have also appeared in the physics jour- of literature dealing with each of the three
nals. In order to construct a meaningful phases of targets differs greatly. The litera-
review, the literature must first be reduced to ture on solids is easily the largest, and so the
a subset. In this instance, LIBS in analytical review of solids is taken almost entirely from
chemistry is chosen. publications within the last 5 years.
This review will cover fundamental stud- Literature on laser-induced breakdown
ies and analytical results and applications of in gases is the oldest. Many of these publi-
LIBS related to the field of analytical chem- cations are in the physics literature and deal
istry. The review is divided according to with mechanisms of breakdown and plasma
target phase. Solids, liquids and gases are diagnostics. There are a smaller number of
treated in sections devoted to each. Articles papers that report upon the determination
from the physics literature are included when of trace metals in gas phase matrices. Most
they are of interest to the analytical chemist. of these are intended to prove the useful-
Spectroscopic studies of laser-induced plas- ness of LIBS in hazard monitoring. In the
mas created in pulsed-laser deposition (PLD) section on gases, the analytical chemistry

1O4@8347/97/$.50
0 1997 by CRC Press LLC
257
literature over the past 10 years is reviewed In 1993,Ibrahim and Goddard published
and appropriate additions are made from “An Overview of Laser-Induced Breakdown
the physics literature. Spectros~opy.”~ They concentrated on the
The section dealing with liquids includes topics of laser-material interaction, local ther-
bulk liquids, isolated droplets, and aerosols modynamic equilibrium, and plasma diag-
generated from liquids. Among the chemis- nostics. The instrumentation was described,
try journals, this literature is the smallest. the use of gated detection was explained,
The physics is of interest primarily because and a few applications were presented. It
of medical applications of laser-induced was not a literature review.
breakdown, and publications which are In a 1993 review by Darke and Tyson
judged to be of interest to the analytical titled “Interaction of Laser Radiation with
chemist have been included. All publica- Solid Materials and its Significance to Ana-
tions related to the elemental analysis of lytical Spectrometry,” applications of LIBS
liquids by LIBS during the last 14 years are were reviewed in the section on laser abla-
included. tion.6 In this paper, LIBS was referred to as
Previous reviews on this topic begin with laser microprobe optical emission.
a 1984 review by Adrain and Watson titled The following year application of LIBS
“LaserMicrospectral Analysis: A Review of to process control was briefly reviewed by
Principles and Applications.”’ This was fol- No11 et al.’ The authors discussed fundamen-
lowed by Cremers and Radziemski who tals of plasma formation and laser-material
published “Laser-induced Breakdown Spec- interaction with respect to optical and heat
troscopy: Principles, Applications, and In- penetration, laser energy, density, and ab-
strumentation” in 1990.2This paper included sorbance. The use of fiber optics and on-line
a brief review of theory and instrumentation sampling was discussed, and a periodic table
for LIBS followed by a few industrial appli- was presented showing elements which had
cations. In 1992,Thiem et al. reviewed LIBS been determined and their detection limits in
theory as part of their paper titled “Lasers in iron ore. This idea was incorporated into the
Atomic Spectroscopy: Selected Applica- review that follows. Each section was con-
t i o n ~ . ”The
~ section on laser-induced plasma cluded with a periodic table or tables with
covered laser-material interaction, plasma elements that have been determined and
production factors (wavelength, energy, and shaded according to detection limit.
properties of target), and emission factors Most recently, LIBS was briefly reviewed
(temperature and electron density). as a section in the comprehensive review of
Majidi and Joseph published “Spectro- atomic emission by Sharp et al. published in
scopic Applications of Laser-Induced Plas- 1995.8
mas” in 1992.4 This publication has perhaps
had the most influence in terms of style and
content on the review that follows. Majidi II. SOLIDS
and Joseph reviewed analytical results on
solids, liquids, gases, and mixed phase sys- A. Fundamental Studies
tems for the years 1987-1992. Emphasis was
on applications of LIBS such as determina- Many papers in this area stem from the
tion of hazardous elements in air, toxic ele- need to more fully understand the LIBS
ments in wastewater, and elements of inter- plasma in order to obtain useful quantitative
est in coals and iron ores. The authors stressed results. One such paper is the 1988 publica-
the fact that LIBS required only optical ac- tion by Chen and Yeung who used the acous-
cess to samples. tic signal generated by a laser-induced plasma

258
 as an internal ~tandard.~ They reported that melting of the solid and included parameters
the magnitude of the acoustic wave was pro- such as thermal diffusivity, pulse duration,
portional to the emission signal for major and density. The second part of the model
and minor elements within the solid target. described the plasma expansion into a
Furthermore, they found that this propor- vacuum.
tionality was independent of laser power and Mason and Goldberg have characterized
focus spot size. a laser plasma in a pulsed magnetic field.14
In the same year, Wood et al. studied the The first part of the paper included spatially
effect of laser pulse duration on soft X-ray resolved emission studies. The authors found
emission from a tantalum target.l0 Using a that when a pulsed magnetic field was ori-
colliding-pulse mode locked dye laser at 620 ented normal to the laser beam, it caused
nm with pulse durations of 100 and 600 fs radial compression and axial expansion in
and a Nd:YAG laser at 1064nm with a pulse the plasma. The emission intensities of both
duration of 70 ps, the plasma emission in the atoms and ions were also increased in this
range from 10to 7 1 m was observed. It was magnetically confined plasma.
found that longer pulses gave relatively more In the second part of the paper, the au-
short wavelength emission and longer emis- thors discuss time resolved emission and
sion lifetimes than short pulses. absorption studies. They concluded that the
In 1989,Coche et al. used laserenhanced increase in emission seen in the plasma was
ionization detection in a laser plasma to study due to Joule heating caused by the induced
the processes of ionization and recombina- secondary current in the plasma. This was
tion.” They used a N, laser (337.1 m, 5 mJ, evident by the fact that the increase in emis-
10 ns) to ablate a solid target. A dye laser sion was seen later in time than the maxi-
was used to selectively ionize species in the mum of the applied magnetic field. Also, the
plasma at different delays relative to the intensity increase was attributed to increased
ablation pulse. Optogalvanic detection was atomization efficiency and longer residence
used to give an indication of the number of times in the plasma.
atoms in the probe volume at the chosen The effects of buffer gas type on the
time delay. In this way, ionization and re- plasma produced by a Nd:YAG laser (1064
combination rates could be inferred. nm, 100 mJ, 7 ns) on a metal target was
Also in 1989, Iida studied the atomic studied by Owens and Majidi in 1991.” They
emission characteristics of a laser plasma in observed an increase in the ratio of A1 WAl
reduced pressure argon.I2Using a ruby laser I intensity in helium gas relative to argon gas
(1.5 J, 20 ns), a plasma about 10 mm in and air. This increase was attributed to the
height, and more than 100 ps in duration was ability of excited helium atoms to transfer
formed at pressures between 0 and 50 Tom. energy to a similar energy level in the alumi-
The plasma at 50 Ton had greatly increased num ion.
line emission and background due to con- More metal target studieswere performed
finement by the Ar atmosphere. The plasma by Lee et al. in 1992.16These authors used
at lower pressure showed less background an ArF laser (193 nm, 100 d,10 ns) to
and less line emission. produce plasmas on copper and lead. The
Another investigation of plasma expan- lead plasma was much larger than the copper
sion was done by Balazs et al. in 1991.13 plasma (5 mm vs. 2 mm) and had a slightly
These authors investigated ruby laser pulses lower excitation temperature as determined
on a copper target, and constructed a two- by Boltmann plots. The temperature of the
part model for the interaction. The first part copper plasma was 13200 to 17200 K, the
of the model dealt with the heating and lead plasma was 11700 to 15300 K.

259
 In 1992 Marine et al. studied plasma In 1994, Kagawa et al. used a XeCl
expansion by optical time-of-flight measure- laser (308 nm, 15 to 70 mJ, 20 ns) to pro-
m e n t ~ . They
'~ determined that the velocity duce plasma on a Zn target in vacuum.21
distribution of ions produced by a UV pulse Time-resolved studies showed a number
of several nanoseconds duration was broad density jump that represented the blast wave
and not well defined. In contrast, the ions expansion into the observed volume. This
produced by a picosecond IR pulse traveled outermost portion of the plasma was shown
with a velocity inversely proportional to the to be ideal for analytical measurements
square root of their mass. Neutral atoms still because the background in this area was
had a poorly defined velocity distribution. greatly reduced.
These authors also noted the appearance of In 1994, Tambay and Thareja studied
a bi-modal temporal profile for YO+ emis- emission in a laser plasma of Cd metal vapor
sion and attributed this to two possible formed in a heatpipe.22They showed that
mechanisms for formation of YO+. emission from the vapor was stronger when
Kuzuya et al. studied the effect of laser the plasma was formed on a tungsten target
energy and atmosphere on the emission char- than it was when the plasma was formed in
acteristics of laser-induced plasmas.18They the gas itself. The authors claimed that this
used a Nd:YAG laser at 1064 nm and pulse was due to pumping of the vapor by soft
energies from 20 to 95 mJ in atmospheres of X-rays formed on the tungsten target.
He, Ar, and air from 1 Torr to 1 atm. The Transition probabilities of 28 Si ion lines
authors reported that maximum emission were determined using a laser-induced
intensity was observed at 95 mJ in 200 Torr plasma as a source by Blanco et al. in 1995F3
of Ar. However, maximum signal to back- Using a Nd:YAG laser (1064 nm, 280 mJ,
ground was obtained in helium at 40 Ton 10 ns) to produce a plasma on pure silicon,
and 20 mJ of power. Images showed the these authors observed emission in Ar and
different sizes and shape of these plasmas. Kr atmospheres. The plasma produced was
Okana et al. studied mass removal in found to have an excitation temperature of
non-metallic inorganic solids and determined 20000 K and an electron density of 1017
relationships between laser power and atom ~ m - Absolute
~. transition probabilities for
~ie1ds.l~ Their paper described vacancy ini- the Si ion lines were calculated.
tiated laser ablation as a process by which Jensen et al. published mechanistic stud-
weakly bound atoms were released from ies of laser-induced breakdown on model
around vacancies and vacancy clusters. The environmental samples in 1995.24They used
atom yield was determined to be an expo- a KrF laser (248 nm, 30 ns) to produce plas-
nential function of laser fluence. mas on SiO, containing Eu and Cr, which
Time resolved emission studies from a were added as the solids Eu20, and &Cr207
laser plasma on sodium chloride were re- to the sand. Detection limits of 100 ppb for
ported by Yago et al. in 1993.20They used a Eu and 2 ppb for Cr were reported. A sample
Nd:YAG laser of 150 ns duration focused was also prepared in which the source of Cr
onto a NaCl pellet. The emission spectra was a solution which was added to the sand
showed self-reversal in air but not in a and then evaporated. This method of sample
vacuum. The plasma was divided into two preparation gave an order of magnitude less
zones, a hot core behaving as an emission signal and a different temporal profile for
zone, and a low-temperature periphery be- the Cr emission.
having as a reabsorption zone. The plasma In 1995, Tasaka et al. studied the emis-
expansion rate was shown to be determined sion of a laser-produced plasma on graph-
by ambient gas pressure. ite.25They used a Nd:YAG laser in He and

260
 in air to form a plasma in which they ob- of different focal length lenses and different
served a “triple plume” composed of three sampling geometries on plasma formation
distinct regions. The authors claimed that and location of species within the plasma.
these regions appeared because of three dif- In 1996, Mulatari et al. obtained time
ferent speeds in the expanding plasma. The resolved images from a laser plasma formed
fastest region was composed of carbon ions on a sample at non-normal incidence.34 They
from the target and N and 0 ions when the varied the angle of the incident laser beam
experiment was done in air. The second with respect to the target from perpendicular
plume was composed of the compressed to nearly parallel. It was found that the plume
neutrals in the vicinity of the shock wave. was generated in the perpendicular direction
The slowest plume was then the target vapor regardless of the laser angle. However, the
composed of larger molecules. Thareja et al. maximum emission signal was obtained with
also studied graphite plasmas at low pres- normal incidence.
sure and found similar temporal profiles.26 Mulatari and Cremers also published a
Intense emission from the CN radical second study that year reporting on the use
has been observed in plasmas produced on of an acousto-optic tunable filter to capture
graphite. With the use of a high-resolution spectrally resolved images.35 They used a
spectrometer, several authors have resolved series of different lenses to examine the dif-
the vibrational and rotational structure in the ferent distributions of elements within each
CN emission bands. The emission from the plasma. By collecting light from the outer-
violet band of CN around 388-nm has been most edge of the plasma, they were able to
used to calculate vibrational and rotational use ungated detection to obtain analytically
temperatures in the laser p l a ~ m a . ~ ~ - ~ O useful spectra with low background.
In 1996, Al-Wazzan et al. studied three- Nemet and Kozma studied the shape of
dimensional number densities of species in emission lines produced on gold targets at
laser produced plumes.31They used absorp- different delays relative to the ablation la-
tion of an expanded dye laser beam to form ~ e r The
’ . ~406.51
~ and 389.79 nm lines were
shadow images of Ba ions in a plume pro- observed. These lines were asymmetric and
duced by excimer laser ablation at 248 nm. shifted relative to their natural wavelength at
They also used fluorescence from planar times up to 1000 ns after the laser pulse.
slices of the plasma to obtain sequential cross- These lines could be described by asymmet-
section images that could be built into a ric Lorenz-type profiles. After 1000 ns, the
three-dimensional image. lines appeared to be very close to their natu-
Al-Wazzan et al. also carried out an ex- ral wavelength and were Lorentzian and
periment in which they observed plasma in symmetric.
vacuum and in ambient oxygen.32In oxygen, In 1997, Kurniawan and Kagawa used a
the expanding plume showed increased tem- long pulse Nd:YAG laser to produce plasma
perature and electron number density at the on a brass target in vacuum.37 The authors
shock front due to increased collisional ex- were especially interested in the secondary
citation rates. In a vacuum, enhancement at plasma formed by compression in the vicin-
the shock front was not observed. ity of the shock wave. The emission in this
Bulatov and Liang obtained full spectra secondary plasma was captured, and it was
at each pixel in the image of a laser-induced observed that when a wedge of aluminum
plasma as depicted in Figure l.33They used was placed very close to the target, the emis-
this technique to create classification maps sion signal increased in the vicinity of the
that gave location of any species of interest wedge. The authors attributed this effect to
within the plasma. They studied the effects increased compression provided by the shock

261
 \
I

262
wave interaction with the immovable alumi- stainless steels. At a laser power of 100 mJ,
num wedge. the detection limit for carbon was 65 ppm
In 1997 Martin et al. used a laser-in- and the RSD was 1.6 %.
duced plasma to establish spectral calibra- Also in 1992, Hader used LIBS for on-
tion of their detector.38 A Nd:YAG laser line quality control of rubber mixing!’ The
(1064 nm, 350 mJ) was focused onto targets author called the technique “remote laser
of ceramic, polyethylene, ZnS, and alumi- microanalysis” and used the acronym
num, all in vacuum. Line pairs in the deep RELMA. The existing technique for quality
UV were used with known transition prob- control of rubber mixing involved a discrete
abilities to determine the relative efficiency sampling step in which a small portion of
of the detection at each wavelength. rubber was taken from a batch and analyzed.
Most recently, Granse et al. modeled This led to problems when it became
laser-induced plasma and compared their neccessary to distinguish between bulk com-
model to experiments with different lasers position fluctuations and inhomogenieties;
and different materials.39 The model they with on-line sampling by LIBS in a number
derived accounted for fluid dynamics of the of randomly chosen positions in the rubber
plasma, absorption of laser energy via in- mix, the two problems could be distinguished.
verse bremsstrahlung, and the dynamics of Analytes were elements found in the cross-
ionization and recombination. linkers and plasticizers added to the raw
rubber.
Lorenzen evaluated other on-line appli-
B. Analytical Results and Applications cations of LIBS in the same year.42 This
time the technique was referred to as “la-
The number of papers describing ana- ser-induced emission spectral analysis” and
lytical results of LIBS studies on solids is given the acronym LIESA. The authors
easily larger than the number of papers deal- described applications such as the determi-
ing with either liquids or gases. For this nation of minor elements in liquid steel,
reason, only publications after 1992 are con- and depth profiling of layers on metallic
sidered in this section. Figures of merit given substrates.
for specific determinationsdiffer greatly from In 1993, Sabsabi et al. used a KrF laser
one author to the next. In the periodic tables (248 nm, 100 mJ) to analyze aluminum al-
which follow this section, the lowest pub- l o y ~ ?They
~ carefully optimized the delay
lished detection limits for the elements in the time by using a PMT to look at the temporal
matrix of interest are used to determine the evolution of several different emission lines.
shading. An aluminum line was used as an internal
Carbon content in steel was determined standard and magnesium was determined at
by Aguilera et al. in 1992.40The authors a few ppm.
observed the emission from the 193.1 nm The following year, Thiem et al. inves-
carbon line because the 247.9 line had an tigated LIBS of alloy targets.44They deter-
interference. The experiment was done in mined Al, Cu, Fe, Ni, and Z n using a
a CO, free environment, and a neighbor- Nd:YAG laser at 532 nm in a vacuum cham-
ing iron line was used as an internal stan- ber. Using non-resonant lines for all the el-
dard. A Nd:YAG laser at 1064 nm was used, ements, the authors were able to generate
and it was found that the slope of the calibra- linear calibration curves from a few hundred
tion curves generated decreased with increas- ppm to a few percent for each element. For
ing laser power. Also, there was a slight the determination of these elements at the
difference in slope for stainless and non- percent level, the authors have found many

263
potential emission lines i n the spectral win- experiment.48A Zn/Ni coating between 2.7
dow from 300 to 400 nm. and 7.2 pm thick on a steel substrate was
Thiem and Wolf used LIBS to analyze analyzed, and a calibration curve of signal
mining ores and compared the results to duration with thickness was found to be lin-
those obtained by a digestion ICP-AES ear. The depth resolution was far poorer than
method.4s Al, Ca, Cu, Fe, K, Mg, Mn, Si, in the mass spectrometric experiment of Allen
and Ti were determined in aluminum and because the energy density at the target was
manganese ore. The comparison showed much greater. Nonetheless, Sn coatings of
that both methods gave roughly the same less than 1 pm, and Cr coatings on the order
accuracy, but the ICP technique had better of a few nm on steel could be determined by
precision. The authors also pointed out the this technique.
similarity in cost for the two techniques, Arnold and Cremers used LIBS to deter-
about $100 K for either. mine metal particles on air sampling filters.49
Investigations of the use of resonant T1 was collected on filter paper by passing
wavelengths in laser ablation had been contaminated air through a filter or by wip-
made as early as 1992 by Borthwick et al. ing a filter on a T1 surface. The laser beam
46 Although most of the work was carried from a Nd:YAG laser at 1064 nm was formed
out with time-of-flight mass spectrometry, into a line at the focus by a pair of cylindri-
the paper is included because of its rel- cal lenses. The detection limit for Tl was 40
evance to LIBS. The authors noted that ng/cm2 of filter paper, and the calibration
when an ablation laser was scanned through curve was linear up to 40 pg/cm2.The 535.05
specific wavelengths, an enhanced ion nm T1 line was used for detection.
yield was detected. More specifically, an Sattmann and Sturmalso investigated the
enhancement in Ga ions in the ablation of use of a multiple Q-switch Nd:YAG laser
GaAs, and an enhancement in A1 ion yields for analysis of steel samples.50Single, double,
in ablation of steels were noted. The ion and multiple pulses were used to produce
yields were even more pronounced when plasma on a low-alloy steel. Material abla-
grazing angles of incidence of the ablation tion, emission intensity, electronic tempera-
laser were used. ture, and electron number density were de-
In 1995, Allen et al. employed resonant termined for each plasma type. All of these
wavelengths in LIBS of thin films."' The parameters were greater for the double and
authors used resonant laser ablation time-of- multiple pulses than for the single pulses
flight mass spectrometry to investigate cop- presumably because of a smaller shielding
per thin films on a silicon substrate. Using a effect in these cases. Calibration curves were
XeCl excimer laser pumped dye laser (5 mJ correspondingly steeper.
pulse energy at 463.51 nm) multi-photon Davies and Telle used a 100 m fiber
ionization of copper was observed. By using optic pair to perform remote LIBS on fer-
an unfocussed laser continually and watch- rous target^.^' A pair of 550 pm (OD) fiber
ing the decay of the copper signal, the au- optics were used to transmit laser light to the
thors were able to determine the thickness of target and to return plasma emission to the
A.
the film between 20 and 100 It was cal- spectrometer. Detection Iimits of 200 ppm
culated that between l e 3and lo-*A per shot or less were found for Cr, Cu, Mn, Mo, Ni,
was removed by the laser, although this was Si, and V. A number of potential emission
at a power below the breakdown threshold, lines to use for calibration were, given for
so emission could not be collected. these elements.
Anderson et al. used LIBS for depth pro- Cremers et al. used a fiber optic probe to
filing without the resonance feature of AIIen's determine Ba and Cr in A singIe 1.5-mm

264
 fiber was used both to deliver laser light and Sabsabi and Cielo once again analyzed
collect emission. A glass plate was used to aluminum alloys by LIBS in 1995.57The
reflect the emission to the spectrometer. The authors characterized the laser plasma on
Ba ion line at 493.41 nm and the Cr atom aluminum targets using Stark broadening
line at 425.44 were used for calibration which of A1 ion lines to determine electron num-
was linear over 4 orders of magnitude for ber densities and using a series of Fe atom
both elements. The detection limit for Ba lines to make Boltzmann plots to determine
was 26 ppm, and the detection limit for Cr electronic temperature. Calibration curves
was 50 ppm. for Mg, Mn, Cu, and Si were constructed
Cremers also determined at what dis- and detection limits were as low as 10 ppm,
tance LIBS could be performed using con- for Mg.
ventional optics rather than fiber optics.53 Soil was analyzed by Ciucci et al. in
By using a beam expander to increase the 1996.58They used both a Nd:YAG laser at
diameter of the laser beam and a pair of 1064 nm and a XeCl laser at 308 nm. The
lenses with adjustable distance to focus this authors found that the background decay
expanded beam onto a target, he was able to in the laser plasma was faster for the 308
produce a plasma and collect light at a dis- nm pulse than for the 1064 nm pulse and
tance of 24 m. Detection limits in a simu- used a correspondingly shorter delay time
lated moon rock were at the level of a few with the XeCl laser. Cu, Pb, and Cr in soil
percent due mostly to the small solid angle were determined, and an entire spectra from
of collection at this distance. 350-700 nm was shown with lines identi-
Bescos et al. again analyzed aluminum fied for the geological survey soil sample
samples in 1995.54 The authors simulta- GXR-2.
neously determined Mg, Mn, Fe, and Pb in Ernst et al. used LIBS to determine Cu
aluminum using a spectral window from 380 in A533b steel.59The first attempt was car-
to 410 nm. Detection limits were around 100 ried out with fiber optic delivery of the
ppm and calibration curves were linear up to laser. This method would have been sig-
about 1%. nificant because in the hazardous environ-
Gonzales et al. also analyzed steel ment of a nuclear reactor pressure vessel,
samples.55Their 1995 publication dealt with Cu concentration in the steel is an indicator
sulfur determination. They reported a detec- of radiation embrittlement and of expected
tion limit of 700 ppm using the 180.73 nm material lifetime. Because the fiber optic
atom line. The analysis was done in a N, could not deliver enough power to produce
atmosphere, and the Fe ion line at 186.47 sensitive detection, beam delivery was done
was used as an internal standard. Precision with conventionaloptics. The Cu line at 324.75
was 7%. was used for calibration, and by using a
Hakkanen and Korppi-Tommola used second-order calibration curve Cu could be
LIBS to study elemental distributions of determined in the range between 100 ppm
paper coatings.56An XeCl laser at 308 nm and 5%.
was used to generate a plasma on paper, and Geertsen et al. revisited LIBS for alumi-
Ca or Si was used as an internal standard. A num samples in 1996.@Using a Nd:YAG at
spatial resolution of about 250 pm could be 1064 nm and a pulse power of 230 mJ the
achieved with a laser pulse power of 200 mJ. authors formed craters -5 l m deep in alumi-
It was estimated that 2 ng of paper coating num with a single shot. The detection limits
were vaporized per shot. Al, Si, Mg, Ca, and for Mg and Cu were 4 and 40 ppm, respec-
C were were monitored quantitatively in a tively, with a precision of about 8%. Spatial
spectral window from 220 to 290 nm. resolution was assessed by rastering the beam

265
over a sharp AI-Cu junction. The best spatial tions, including Hg in air (5 ppm detection
resolution obtained was 6 pm. limit), and pollutants in powerplant smoke
Marquardt et al. determined Pb in paint and soil." These authors analyzed the geo-
using a Nd:YAG laser at 532 nm coupled graphical survey GXR-2 sample and deter-
into a fiber optic probe.6l The probe con- mined concentrations for 18 elements.
sisted of two fibers, excitation and collec- Vadillo and Laserna analyzed geologi-
tion, terminated in a probe head with an cal samples of vanadinite, pyrite, garnet, and
aspheric lens to focus the laser light to a spot quartz.65These determinations were done in
and focus plasma emission into the collec- vacuum to increase the lifetime of the ionic
tion fiber, The common end of this fiber is species and of the ion lines observed. Fe,
shown in Figure 2. Different combinations Mn, Mg, and Si were determined in each of
of excitatiodcollection fiber diameters were the rocks, and A1 was also determined in
tried. The detection limit for Pb was 140 garnet. Differences in composition were as
ppm, and precision was 5 to 10 % even when expected.
analysis was carried out through layers of Laserna et al. used LIBS to analyze the
non-lead containing paint. surface of solar cells.& A N, laser at 337.1
Miziolek also reported on a LIBS probe nm was focused onto the cell and the cell
in 1996.62This application was for determi- was rastered under the focus providing spa-
nation of heavy metals in soils, and the probes tial resolution of about 30 pm. The C ion line
were to be used in a cone penetrometer truck. at 588.9, Ag atom line at 546.5, Si atom line
In one probe, a compact laser was mounted at 634.7, and Ti atom line at 625.9 were used
in the probe head. In the other probe, laser in determinationsof these four elements. The
light was carried through a fiber optic to the concentration of each element was mapped
probe head. The compact laser head pro- across the surface of the solar cell.
vided the more sensitive probe with detec- In another application of LIBS to alloys,
tion limits for Pb, Hg, Cr, Cd, and Zn of 1 to Kim et al. determined aluminum in a zinc
10 ppm in sand and silt. The fiber optic The authors used a Nd:YAG laser
delivery provided less power but a higher (1064 nm, 105 mJ, 3 ns) in vacuum, air, and
laser repetition rate; 15 mT vs. 28 mJ, 30 Hz argon. The presence of A1 in the Zn alloy
vs. 113 Hz. was important because of its detrimental
Also in 1996, Cremers et al. introduced effect on its welding properties. The delay
a portable LIBS instrument and used it to and the distance from the surface of the tar-
determine hazardous elements in soil, in get were optimized in each atmosphere.
paint, and in samples on filter paper.63A Because the A1 atom line at 308.22 nm was
compact laser head was used as the probe. used for calibration, large delay times were
Ba, Be, Pb, and Sr were determined in soil used to ensure no contribution from a shorter-
with detection limits of 265,9.3,298, and 42 lived Zn interference. In air, a 30 ys delay
ppm, respectively. In paint, the detection limit was used and in Ar a 50 ps delay was used.
for lead was much greater because the 405.8 The Zn line at 307.59 was used as an internal
nm line could not be used due to spectral standard.
interference. On filter paper, Be and Pb were In 1997, Maravelaki et al. used LIBS to
determined with detection limits of 21 ng/cm2 monitor the laser cleaning of marble arti-
and 5.6 FLg/cm2,respectively. In the filter Although lasers had been used previ-
paper studies, some particle size effects were ously for the cleaning of such treasures, these
evident. authors were the first to examine the emis-
Palleschi et al. used a 400 mJ Nd:YAG sion from the plasma. Crusts of 20 to 600 pm
laser at 1064 nm for a variety of determina- consisting of gypsum, iron oxides, soot, and

266
 f/2 Lens

Collection Fiber

Excitation Fiber

Aspheric Lens

FIGURE 2. Common end of fiber optic LlBS probe used by Marquardt et al. for determination of lead in paint.
calcite were ablated from the marble by a laser (532 nm, 10 ns) was focused onto drop-
Nd:YAG laser. The plasma formed in this lets generated by a Berglund-Liu vibrating
ablation was analyzed spectroscopically to orifice aerosol generator. Breakdown thresh-
determine the endpoint of the cleaning. olds were determined on aerosols of differ-
Figures 3 and 4 show graphically the ent size generated from liquids of various
detection limits for elements that have been refractive index, density, surface tension, and
determined using LIBS in sand, ore, or soil chemical structure. Thresholds were seen to
and in steel or alloy. decrease with droplet size. Refractive index
had no apparent effect. Density, surface ten-
sion, and chemical structure had effects that
111. LIQUIDS could not be independently examined.
In 1987, Hsieh et al. used a Nd:YAG
A. Fundamental Studies laser at 532 nm to examine the exact lo-
cation of breakdown initiation on water
Besides papers in the field of analytical droplets generated by a Berglund-Liu vi-
chemistry, there have also been a number of brating orifice.72They found that as the en-
papers that have dealt with the physics of ergy of the laser increased, the breakdown
laser-induced breakdown in aqueous solu- moved from outside the droplet to the inside.
tion. Most of these deal with the processes of Theoretical calculations showed that the
cavitation and the factors that can affect the curved liquid gas interface of a water droplet
breakdown threshold. in air focused laser energy at one point just
In 1984, Armstrong identified three time inside the illuminated face and at another
domains in the heating of an aerosol: the point just beyond the shadowed face.
acoustic regime, the internal conductive re- The effect of laser wavelength and irra-
gime, and the external conductive regime.69 diance on spectra from laser-induced break-
He explained the phenomenon of aerosol- down of single levitated aerosol droplets was
enhanced air breakdown as being caused by examined by Biswas et al. in 1988.73Drop-
increased electron collision frequency in the lets of glycerine saturated brine solution
hot, dense, vapor surrounding the heated approximately 18 pm in diameter were opti-
aerosol particles in the beam. cally levitated and probed with a Nd:YAG
Three years later Armstrong used a CO, laser at 1064, 532, and 355 nm. Emission
laser at 10.6 pm to explode aerosols gener- lines of NayC, and N were observed. It was
ated by a vibrating orifice aerosol genera- shown that the energy required for break-
tor.7O A UV laser was used to produce shadow down, and the time of plasma emission
images of the exploding droplet, and a phase increased with increasing wavelength. The
Doppler particle analyzer system deter- optimum time delay for determination of
mined size and velocity of the expelled par- these elements increased with laser wave-
ticles at a distance of 2 mm from the original length and showed more variance from ele-
drop. It was found that the CO, laser power ment to element at longer laser wavelength.
affected the size of particles produced in the Also in 1988, Zheng et al. acquired tem-
breakdown for original drop sizes between porally and spatially resolved spectra of la-
30 and 50 pm. The speed of the particles ser-induced breakdown of a 40 pm 4 M NaCl
produced was independent of laser power. droplet using a Nd:YAG laser at 532 nm.74
Chylek et al. studied the effect of size Using fiber optic ribbons and a streak camera,
and material of liquid aerosols on break- these authors observed that emission began
down thresholds in 1986.71The setup that first at the shadowed face of the droplet and
they used is shown in Figure 5. A Nd:YAG propagated toward the illuminated face.

268
 Ores, Rocks, and Solids

Fe
-

Ru

0s Po At
YF-! Rn

Key: 0 c 10 ppm
10 - 100 ppm

FIGURE 3. Elements determined in ores and soils shaded according to detection limit.
 Metals
-
H
-
Li Be
-
Na
I

I I I I I I I I I

Fr

Key: 0 ~ 5 ppm
0
50 - 200 ppm
rn > 200 ppm

FIGURE 4. Elements determined in metals shaded according to detection limit.

Pyroelectric Detector

a
Microscope Viewing

Frequency Generator

FIGURE 5. Schematic of experimental setup used by Chylek el al. to investigate LlBS of aerosols.
 Hammer et al. also examined thresholds bremstrahlung was responsible for break-
with ultrashort laser pulses.75Using pulses down when nanosecond laser pulses were
of 2.4 ps, 400 fs, and 100 fs from a pulsed used. In this case, the breakdown threshold
dye amplifier, they determined breakdown behaved probabilistically. The probability of
thresholds in saline solution, high purity breakdown scaled linearly with the log of
water, and tap water. The thresholds in these laser power for low powers, and scaled para-
media were not significantly different. With bolically with the log of laser power at higher
the 2.4 ps pulse the breakdown threshold power. In contrast, with laser pulses of less
was found to be 5 x 10" W cm-2, with a 400 than nanosecond duration, multi-photon ion-
fs pulse it was 1.3 x 10l2W cm-2 and with a ization appeared to be the mechanism of
100 fs pulse it was 5.65 x 10l2W cm-2. plasma formation, and a definite threshold
Kitamori et al. used the acoustic wave above which breakdown would occur could
generated during particle-induced break- be determined.
down to detect polystyrene particles in aque- Sacchi returned to this topic in 1996 and
ous solution.76A Nd:YAG laser (532 nm, produced images of laser-induced breakdown
-1 mJ, 6 ns, 10 Hz) was focused into a on solids under water. By using a dye laser
solution. In pure water, the energy density oriented parallel to the underwater target and
produced by this beam was not sufficient to delayed relative to the ablation laser, the
cause a breakdown. When a particle en- authors obtained shadowgraphs of the evolv-
tered the probe volume, however, a break- ing plume and shock wave. A series of im-
down occurred a-nd subsequently an acous- ages followed the growth and decay over a
tic wave was generated. By detecting these total time of about 700 ps. A schematic of
acoustic waves, the probability of break- the setup used is shown in Figure 6.
down was determined. By analyzing solu- In 1992, Pinninck et al. studied the effect
tions of known particle concentration, a of resident particles on laser-induced
calibration curve of breakdown probability breakdown threshold^.^^ It was found that
vs. concentration was generated. This curve even at wavelengths where the particles did
was used in the analysis of solutions of not absorb laser light, there was a decrease
unknown particle concentration. in breakdown threshold in particle laden air
Vogel et al. used a Nd:YAG laser to vs. clean air. This was taken to mean that
produce and image a breakdown in ~ a t e r . 7 ~ particles were focusing laser light. Location
The fundamental wavelength of 1064 nm of plasma formation (inside, shadowed side,
was focused into the water to produce a or illuminated side) was found to depend on
breakdown. A portion of the beam was fre- ionization potential, gas pressure, and laser
quency doubled to produce 532 nm light that wavelength. Refractive index had no effect
was put through a beam expander and al- on breakdown threshold, in agreement with
lowed to pass through the underwater break- the observations reported by Chylek in
down, providing a shadow of the plasma and 1986.71
the cavitation bubble. The size of the plasma In 1993, Nyga and Neu created a plasma
and cavitation bubble were examined for on calcite submerged in water using two
1064 nm laser pulse durations of 30 ps and fiber optics to deliver two pulses from two
6 ns. XeCl lasers.80Both pulses were 30 ns pulses
Sacchi discussed the mechanisms of la- at 308 nm through 600 pm fibers that were
ser breakdown in water and described the brought in close proximety to the target
differences in threshold for short pulse and with no focusing lenses. The pulses were
long pulse lasers in 1991.78 He claimed that separated in time by -300 ps, and the second
the avalanche breakdown process or inverse fiber was equipped with a 308 nm dielectric

272
 I n I

-, &
I
.. .. -"I
d .. .. -
m
0

.. .. a?
. v)
c
a

..
.
E L
6
.. P
m

...
r
m

.. .. .. -
.-Cm
.. rn
..
i05
-
0

... .. P
Q)
v)
3

. .. a
3
c
Q)
m
Iu
..
..
..
. ..U .. I
.. .. ..
~ r n

. . .
* . u-
0

273
beam splitter and used to collect emission When a second spark which was generated
from the calcite. No quantitative results were from a second laser and delayed by 18 ps
reported. relative to the first was used to form the
Feng et al. also examined the effect of plasma, the detection limit for B was re-
laser pulse duration on breakdown thresh- duced to 80 ppm. This improvement was
olds in water.81 In their 1997 paper, they attributed to formation of the plasma within
determined the mechanism of breakdown a cavitation bubble when the RSP method
with nanosecond pulses to be due to cas- was used. The experimental setup is shown
cade ionization, and the mechanism of in Figure 7.
breakdown with ultrashort pulses to be In 1987, Wachter and Cremers reported
multi-photon ionization, similar to Sacchi’s a detection limit of 100 ppm for uranium in
findings.78These authors, however, went solution by LIBS.83The plasma used was
on to model the breakdown phenom- formed by a Nd:YAG laser (1064 nm,
enon in water relating the breakdown 260 ml) on the surface of a 4 M nitric acid
threshold to laser power, spot size, and solution contained in a small glass vial.
pulse duration. They derived an extended Each analysis was done by averaging 1600
non-linear Schrodinger equation to de- laser shots. This averaging was necessary
scribe the relationship. to overcome poor shot-to-shot precision due
mainly to small variations in lens to sample
distance.
B. Analytical Results In 1987, Cremers conducted an experi-
ment to determine the maximum lens to
Quantitative results for LIBS with liq- sample distance (LTSD) that could be used
uid samples are presented in two sections in a LIBS e ~ p e r i m e n tWith
. ~ ~ a 250 mJ pulse
based on the nature of the analysis. The from a Nd:YAG laser at 1064 nm, it was
first section will cover determinations done found that a 2 m focal length lens could
in bulk liquid. This implies that a laser produce breakdown on molten metal. The
spark is generated either on the surface or larger problem was collection of the emitted
beneath the surface of a relatively large light. A fiber optic could collect enough light
liquid sample. The second section covers at 0.5 m from the plasma for modest figures
determinations done in aerosols or droplets of merit to be obtained. At distances farther
generated from liquid samples. Analysis of than this, the sensitivity of the experiment
dry aerosols generated from solids are in- decreased rapidly.
cluded in the last section on gas phase de- In 1993, Aragon et al. used LIBS to
terminations. determine carbon content in molten
They used a Nd:YAG laser (1064,200 mJ, 8
ns) focused onto molten steel in a crucible
C. Bulk Liquid under an argon atmosphere. In order to pro-
vide a homogeneous sample, a jet of argon
In 1984, Cremers et al. reported the gas was directed downward into the cru-
use of a repetitive single spark (RSS) and cible. This served to remove the topmost
repetitive spark pair (RSP) for the deter- layer of molten liquid which would be en-
mination of Li, Na, K, Rb, Cs, Be, Mg, Ca, hanced in the lighter elements. Carbon was
B, and A1 in aqueous and organic soh- determined by ratioing of the C 193.09 nm
tions.82With the single spark, detection lim- to Fe 201.07 nm intensity and a detection
its for all elements except Li were >1 ppm. limit of 250 ppm was found with an RSD
The detection limit for B was 1200 ppm. of 6%.

274
 .....................

-8
Q)

E
0
m

275
 In 1995, Stolarski et al. produced solution. When micron-sized particles of
sub-surface plasmas in saline solution, ErBa,Cu,O,, rather than a soluble salt, were
triple-distilled water, and tap water with used as the source of Er, the detection limits
Nd:YAG laser pulses of <5 mJ and pulse decreased by a factor of lo3. The improve-
durations of 5 ns and 80 The focus- ment illustrated the increased sensitivity of
sing lens used was 2.54 cm. in diameter particle-induced breakdown as compared to
and had a focal length of 17 mm. This breakdown in particle-free solution. This
lens was chosen to approximate the lens experiment also suggested that 500 nm laser
in the human eye. The authors found that light had analytical advantages over 1064 nm
an energy of 1.5 mJ was enough to pro- light.
duce a breakdown at either pulse dura- In 1997, Ho and Ng used both a Nd:YAG
tion, and found that the plasma tempera- laser (532 nm, 12 ns) and an ArF laser
tures and electron number densities (193 nm, 15 ns) to determine Na in aqueous
depended little on pulse duration for the solution.89In order to increase absorbance of
two studied. These authors observed sodium the laser light, methyl violet was added to
emission around 590 nm in a 0.9% NaCl the solution to be analyzed. It was found that
solution, but no quantitative analysis was the 532 nm laser light produced a more vis-
performed. ibly intense plasma; because the continuum
Also in 1995, Ito et al. used LIBS to emission was higher in the 532 nm produced
determine colloidal iron in water.86They plasma, the 193 nm laser light actually gave
focused a Nd:-YAG laser (1064 nm, better detection limits for Na. A detection
100mJ, 10 ns) into a flowing stream of limit of 230 ppb was reported for Na in the
water containing FeO(0H) as submicron methyl violet solution with the plasma formed
particles. The detection limit for iron was by the ArF laser.
in the pprn range. This technique differed In 1996, a similar experiment was per-
slightly from previous techniques in that formed by Paksy et al., who used a Nd:YAG
it relied on particle-induced breakdown (1064 nm, 15 mJ, 4 ns) focused onto the
for the formation of the plasma. A plasma surface of a molten alloy in an argon atmo-
was formed only when a FeO(0H) par- sphere.% Because the laser power used was
ticle was present in the probe volume. significantly less than that used by most
Because of the extremely small size of authors, the delay of the detection relative to
the particles, however, it was essentially the laser was also less. A 100 ns delay and
a determination of iron in water. The fol- 1000 ns gate width were used. Detection
lowing year, the same authors used a sec- limits were 0.001% for Si in Fe, 0.006% for
ond laser delayed by 1 ps relative to the Cr in Fe, 0.06% for Si in Al, and 0.007% for
first in a technique similar to Cremer’s Mn in Al.
repetitive spark pair. With this technique, Arca et al. determined Cr, Pb, and Cu in
they were able to improve the detection aqueous solution with a single laser shot on
limit for iron to 16 ~ p b . ~ ~ the surface of the water.91Detection limits
In 1996, Knopp and Scherbaum used a were 100, 100, and 50 ppm for the three
dye laser at 500 nm with a pulse energy of elements, respectively. Calibration curves
22 mJ to produce a sub-surface breakdown were linear over 1 order of magnitude. The
in aqueous solution.88They reported detec- authors also analyzed aqueous solutions by
tion limits for Cd, Pb, Ba, Ca, Li, and Na of placing a drop of water to be analyzed on a
500 ppm, 12.5 ppm, 6.8 ppm, 130 ppb, KBr pellet. When this technique was used,
13 ppb and 7.5 ppb, respectively. No signal detection limits were improved by almost an
was obtained for either Hg or Er at 0.1% in order of magnitude.

276
D. Aerosols and Droplets In 1992, Ng et al. used an ArF laser
(193 nm, 150 mJ) to analyze an aerosol
In 1983, Radziemski et al. used a nebu- generated by a concentric glass nebulizer/
lizerheat chamber system to produce an spray chamber traditionally used for ICP-
aerosol from a liquid solution and analyzed AES.96 The authors determined the opti-
the aerosol with LIBS?* Detection limits for mum delay for detection to be 6 ps relative
Na, P, As, and Hg were reported to be 6 ppb, to the laser pulse. Detection limits for Na,
1.2 ppm, 0.5 ppm, and 0.5 ppm, respec- Li, In, Al, Ga, Ca, Mg, K, and Sr were
tively. The authors determined that local determined to be 0.9, 0.3, 10, 3, 3, 8, 3, 2,
thermodynamic equilibrium existed in the and 20 ppm, respectively.
aerosol spark for time delays greater than 1 The following year Parigger and Lewis
ps relative to the laser. used a picosecond XeCl laser (308 nm,
In 1987, Eickmans et al. spatially re- 4 mJ) to produce LIB in 66 pm water
solved the emission from laser-induced droplets.97 Plasma diagnostics showed a
breakdown of an aerosol.93Approximately -
temperature of 10 000 K and an electron
45-pm-diameter droplets were formed by a number density of 10l8~ m - The ~ . Ha line
vibrating orifice aerosol generator which was at 656 nm was used as an internal stan-
synchronized with the Nd:YAG laser beam dard and the detection limit for sodium
at 532 nm. Emission of Li, Nay and H was was -1 ppm. Poulain and Alexander per-
observed in 5 M salt solutions of either NaCl formed a very similar experiment in 1995
or LiCl. Spatial resolution o f the emission using a KrF laser (248 nm,200 mJ). The
revealed that in the plasma plume in the detection limit for Na was found to be
vicinity of the illuminated face of the drop- 165 p ~ m . ~ ~
let, emission lines showed Stark broadening Haisch and Paine reported on the
and self-reversal. Near the shadowed face of characterization of colloidal particles by
the droplet emission lines were narrower and LIBS in 1996.99Field flow fractionation
less intense. was used to sort particles by size and
In 1988, Essien et al. used LIBS to deter- deposit them onto filter paper using par-
mine Cd, Pb, and Zn in an aerosol.” A titioned pumping to retain time resolu-
Nd:YAG laser (1064 nm, 100 mJ, 5 ns) was tion. A cylindrical lens was used to focus
focused onto an aerosol generated by a nebu- the laser onto the paper to avoid ablation
lizerheat chamber. Detection limits were 19 of the paper which could occur if the
ppb, 210 ppb, and 240 ppb for Cd, Pb, and energy density became too high. Detec-
Zn, respectively. tion limits for Si were in the ppb range.
Also in 1988, Archontaki and Crouch Figures 8 and 9 show graphically the
used an isolated droplet generator to pro- detection limits for elements that have been
duce equally spaced, uniformly sized drop- determined using LIBS in bulk liquids and
l e t ~ The
. ~ ~experimental setup was such in aerosols generated from liquids.
that a single drop of known size was al-
ways in the probe volume of the laser for
a given repetition rate. A Nd:YAG laser IV. GASES
(1064 nm, 100 mJ) was used to determine
detection limits for Li, Nay Mg, Ca, Mn A. Fundamental Studies
and A1 of 0.3, 2.2, 1.9, 0.4, 7.2, and 5.2
ppm, respectively. Droplet diameter had In 1988, Kumar and Thareja studied
little effect on the detection limit for sizes laser-induced gas breakdown in the pres-
between 58 and 75 pm. ence of an electric field.loOThey used a

277
 t
Iy:

I I I 1 I

.- r3
3 n
U .-E
.I

1
I c I
fn
c
c
a)
E
iij
I Q-0 I cd
w
7 I -
K
3
V - A i?LL

4I0
.. ............
....... .....
......

278
 Liquid Aerosols

/H/

Zr Nb Mo Tc Ru Rh
G++&++ 1Bi Po At Rn

I I

! 1 0 - 100ppm 1

FIGURE 9. Elements determined in liquid aerosols shaded according to detection limit.

XeCl laser (308 nm, 60 mJ, 8ns) fo- that excitation temperature and electron num-
cused between two electrodes separated ber density in the plasma remained essen-
by 6.5 mm. A transverse static electric tially constant with variations in the ambient
field was produced by applying differ- conditions.
ent potentials to the electrodes. In order
to study the effect of the field on break-
down thresholds, the current on the elec- B. Analytical Results
trodes was adjusted so that no break-
down occurred using only the field or In this field, Cremers and Radziemski
only the laser. The authors were then again did pioneering work. In their 1983
able to determine that pre-breakdown publication, they detected chlorine and
electron densities of 1Olo ~ m created - ~ fluorine in air.'" Detection limits for these
by the field were sufficient to lower the two gases were 8 and 38 ppm by weight,
laser energy density needed to produce respectively. These values corresponded
breakdown. to absolute detection limits of 80 and 2000
Parigger and Lewis used a Nd:YAG la- ng. In a helium atmosphere, the absolute
ser (1064 nm, 300 mJ, 7.5 ns) to produce a detection limits improved to 3 ng for both.
plasma in CO.'O' They observed the emis- Experiments were carried out in which mo-
sion of the C, Swan band around 565 nm to lecular gases containing both chlorine and
determine vibrational temperatures in the fluorine were introduced to the sampling
decaying plasma The observed spectra was chamber. The relative intensities of the
fit with a model in which temperature was chlorine and fluorine signal were found to
a parameter. At a time delay of 30 ps rela- be an indication of the number of each of
tive to the laser pulse, a temperature of these atoms in the molecular gas. All ex-
6745°K was obtained. This temperature was periments were carried out using a
confirmed by use of the CN violet band Nd:YAG laser (1064 nm, 100 mJ, 15 ns)
around 388 nm to calculate the vibrational and detecting C1 at 837.6 nm and F1 at
temperature in a similar way. 685.6 nm.
Yagi and Huo investigated breakdown These authors also detected Be in air.105
in H, gas at different pressures initiated by The limit of detection in this experiment was
a KrF laser (248 nm, 250 d, 20 ns).lm This 0.5 ppb by weight. The Be was introduced
work was related to Raman spectroscopy in into a chamber by laser ablation and diluted
that the breakdown threshold determined by air. A second laser produced a break-
the maximum energy with which the Raman down in the chamber from which the emis-
transitions could be pumped. The authors sion was measured. It was found that the
found that below 600 Torr breakdown emission from the Be I1 doublet at 313.1 nm
thresholds were pressure independent. was stronger than any atomic emission even
Thresholds decreased between 600 and 3000 at delay times of 20 ps. This doublet was
Torr, and increased slightly above 3000 used for calibration that was linear over 4
Torr. orders of magnitude.
Yalcin et al. investigated the sensitivity Cremers et al. used LIBS to determine
of a laser spark produced by a Nd:YAG at sodium and potassium in a coal gasifier
532 nm to ambient conditions such as varia- stream as early as 1983.63These elements
tion in background gas, presence of particles, were present as particulates in the stream
and humidity.lo3This was done in order to and so again the question of whether the
assess the potential usefulness of LIBS in target was a gas or solid arises. Regardless,
toxic metal monitoring. The authors found the authors were able to determine Na with

280
 a detection limit of 4 ppb by weight using In 1991, Cheng et al. studied poly-
the doublet at 589 nm. Figures of merit for atomic molecular impurities in helium gas
potassium were not given. In conclusion, at ppm levels.'@ Using ungated detection,
the authors noted that sulfur could also be they determined B,H,, PH,, and ASH, with
monitored. detection limits of 1,3, and 1 ppm, respec-
Sneddon extended this technique to the tively. The ionic phosphorous lines at
determination of P in air and lowered the 602.4, 603.4, 604.3, and 605.5 nm were
absolute detection limit of C1 in air. In a used. As lines were observed at 228.8,
1988 paper, he detected phosphorous at 15 235.0, 278.0, and 286.0 nm. B lines were
pprn and C1 at 60 pprn by weight in air.106A observed at 336.0 and 434.5 nm. The plas-
calculated plasma volume of 0.0 10 cm3was mas were produced by a Nd:YAG laser at
used to determine an absolute detection limit 532 nm.
of 60 pg for C1. Also in 1991, He and co-workers fo-
Ottenson et al. reported on the applica- cused a pulsed laser onto a metal rod in an
tion of LIBS to analyze particulates in a coal atmosphere of helium or argon seeded with
combustion vent in a 1989 publication.lmA a reaction gas to produce ion-molecule
Nd:YAG laser was triggered on the scatter- complexes and observe their emission.l1°
ing from a HeNe laser to explode particles Ion lenses were used to preferentially draw
detected in the probe volume. The experi- ions into the He or Ar gas for reaction with
ment used on-line temperature correction to the seeded gas. Emission from species such
account for different excitation conditions in as A10, AlO-CO, Al-Ar, and A1+H2was
plasmas formed on different size particles. observed.
Moreover, the size of particles to be sampled Joseph and Majidi used an electrother-
could be adjusted by adjusting the laser mal atomizer to create a gas phase sample
power. With lower laser power, only larger from a liquid for analysis by LIBS.ll1A few
particles were sampled, the smaller ones pl sample was deposited in the furnace and
were not sufficient to provide plasma for- was dried, ashed, and atomized. The setup is
mation. The on-line temperature correction shown in Figure 11. At the onset of atomiza-
automatically accounted for differences in tion, a Nd:YAG laser (1064 nm, 100 mJ)
excitation conditions with varying laser was focused into the furnace and fired at a
power. repetition rate of 10 Hz. Emission from the
In 1990, Morris et al. used laser- plasma was collected through the dosing hole
induced breakdown as a detector for gas of the tube. Experiments with cobalt and
chromatography.lo8 The UV laser light cadmium gave absolute detection limits of 5
was focussed into the effluent from a gas and 50 pg.
chromatograph at a power below the Casini et al. also determined several
breakdown threshold of the carrier gas. elements of interest in air using LIBS.l12
When eluting carbon containing molecules Their experiment used a Nd:YAG laser
were present in the effluent, the threshold (1064 nm, 40 mJ, 7 ns) to produce plasma
for breakdown was lowered and a plasma in an atmospheric pressure chamber as
formed. The organic analytes were indi- shown in Figure 12. Using the saturated
rectly detected by optically detecting concentrations of air over chosen liquids at
plasma formation. A schematic is shown the given temperature, and then successively
in Figure 10. No emission from the plasma diluting these concentrations in the cham-
was measured, and so the potential of ber, yielded calibration curves for a number
this technique was only as a universal of elements. C1 was detected at 449.0 nm
detector. with a detection limit of 60 ppm, S at

281
0
9
0

m
3
C

U
-Y
m
e

U
xa
P
3
c.
8
c
0

e
 ai
0
Q
E
7
c

c
0
EQ
!cnz
Q
.-C
0)
m
-I
ti
.-
c

a
z
'0
C
Q
c
Q

0
I
7

s
'0
I7
Q
7
c
a)
v)
c
0
.-c0.
Q

E
c
(2
7
7

W
U
3
rrr
U.

283
Parabolic Mirror Lens
A
Monochromater

Nd:YAG Laser

Laser Controller Computer

FIGURE 12. Shematic of setup used by Cassini et al. for air analysis using LIBS.
 415.3 nm with a DL of 200 ppm, P at 442.1 time the breakdown was produced by a
nm with a DL of 200 ppm, Na at 37 1.1 nm Nd:YAG (1064 nm, 220 mJ, 6 ns) in
with a DL of 110 ppm, Hg at 404.7 nm with hydrogen at 150 and 810 Torr. The H a
a DL of 50 ppm, Be at 38 1.4 nm with a DL line at 656 nm was used to calculate
of 130 ppm, and As at 454.4 nm with a DL electron number densities in the decay-
of 130 ppm. ing plasma with temporal resolution as
In 1994, Flower et al. used LIBS to high as 6 ns early in the plasma and 1 ps
monitor metal aerosol emission generated at later times. Spatial resolution was
by a pneumatic neb~1izer.l'~ A chromium also achieved by focusing different por-
salt solution was passed through the nebu- tions of the 1:l plasma image onto the
lizer and a heating pipe to produce a gas spectrometer entrance slit. Spatial reso-
with known concentration of Cr. The detec- lution was approximately 50 pm. Sur-
tion limit was 200 ng Cr/standard cubic face plots for electron number densities
meter of air (scm). The laser used was a resolved in time and space were then
Nd:YAG (1064 nm, 180 mJ) and detection created. Electron number densities var-
was at 312 nm. ied between 1OI6 and l O I 9 ~ m - ~ .
In 1994, Lazzari et al. detected mercury In 1996, Haisch et al. again used LIBS
in air using LIBS with a Nd:glass laser (400 to detect chlorine in the gas phase.l18 Two
mJ, 8 ns) to produce plasma in a chamber experimental setups were described. The
filled with saturated Hg vapor and succes- first was a benchtop experiment using a
sively d i 1 ~ t e d . lWith
~ ~ this technique, they 320 mJ Nd:YAG laser. The second setup
obtained a detection limit of 10 ppb by ob- used a miniature Nd:YAG laser built into
servation of the 253.7 line. a sensor head. This miniature laser pro-
Zhang et al. performed LIBS in a par- vided 18 mJ of pulse energy. Detection of
ticle loaded methane/air flame and in an C1 in chlorinated hydrocarbons was ac-
oil-fueled combustor vent.115In their 1995 complished by focusing the laser onto a
publication, they described the addition copper target in the presence of the gas to
of a flange with four movable windows to be analyzed. This provided for formation
a preexisting combustor vent. LIBS was of CuCl in the plasma. The luminescence
performed through these windows which of the D-system of CuCl was then detected
were kept clean by a flow of N, gas and around 440 nm. Detection limits were a
could be switched out periodically for few ppm.
cleaning. Figure 13 shows the elements determined
In 1995, Nordstrom studied the laser- in a gas phase matrix and their relative de-
induced plasma emission spectra of N,, 0,, tection limits.
and ambient air from 350 to 950 This
range was covered with approximately 0.5
nm resolution and atomic, ionic, and mo- ACKNOWLEDGMENTS
lecular emission from a plasma created by a
CO, laser (10.6 pm, 320 mJ, 200 ns) was The authors would like to acknowl-
observed. Spectra were computed for N, 0, edge the financial support of the Engi-
and a few molecular species, and a compari- neering Research Center (ERC) for Par-
son between experiment and calculation was ticle Science and Technology at the
done. University of Florida, the National Sci-
The same year, Parigger et al. again ence Foundation (NSF) grant #EEC-94-
published a fundamental paper on laser- 02989, and the Industrial Partners of the
induced plasma in the gas phase.117This ERC .

285
N
m
Q) Gases

H He
-
Li Ne
-

Na Ar
-
K Ti V Kr

Rb Zr Nb Xe

cs Hf Ta Rn
-
Fr ~ Ra Ac

Key: 0 c 10 pglscm
10 - 100 pg/scm
> 100 pg/scm

FIGURE 13. Elements determined in gas phase matrices shaded according to detection limit. The unit scm refers to a standard cubic meter of gas.
REFERENCES 19. Okano, A., Matsuura, A. Y., Hattori, K.,
and Itoh, N., J. Appl. Phys., 73,3158,1993.
1. Adrain, R. S. and Watson, J., J. Appl. Phys.
20. Yago, H.,Furuta, K., Ishikawa, K., and
D, 17,1915,1984.
Komura, H., Phys. Stat. Sol., 179,223,1993.
2. Radziemski, L. J. and Cremers, D. A,, SPIE,
21. Kagawa, K., Kawai, K., M.Tani, and
1318,71,1990.
Kobayashi, T., Applied Spectroscopy, 48,
3. Thiem, T. L., Lee, Y., and Sneddon, J., 198,1994.
Microchemical Journal, 45,1, 1992.
22. Tambay, R. and Thareja, R. K., Laser
4. Majidi, V. and Joseph, M. R., Critical Re- Chemistry, 14,225,1994.
views in Analytical Chemistry, 23,143,1992.
23. Blanco, F., Botho, B., and Campos, J.,
5. Ibrahim, A. and Goddard, B. J., Jurnal Physica Scripta, 52,1995.
Fizik Malaysia, 14,43,1993.
24. Jensen, L. C., Langford, S. C., Dickinson,
6. Darke, S. and Tyson, J., J. Analytical Atomic J. T.,and Addlman, R. S., Spectrochimica
Spectroscopy, 8, 145,1993. Acta, 5OB,1501,1995.
7. Noll, R., Sattmann, R., and Sturm, V., SPIE, 25. Tasaka, Y., Tanaka, M., and Usami, S.,
2248,50,1994. Japanese Journal of Applied Physics, 34,
1673,1995.
8. Sharp, B. L., Chenery, S., Jowitt, R., Fisher,
A., and Sparkes, S. T., J. Analytical Atomic 26. Thareja, R. K. and Dwivedi, R. K., Laser
Spectroscopy, 10,139,1995. and Particle Beams, 13,481,1995.
9.Chen, G. and Yeung, E. S., A-nalytical Chem- 27. Hatem, G., Colon, C., and Campos, J.,
istry, 60,2258,1988. Spectrochimica Acta, 49A,509,1993.
10. Wood, 0.R.,Silf'vast, W. T.,Tom,H. W. K., 28. Dreyfus, R. W., Kelly, R., and Walkup,
Knox, H., Fork, R. L., Brito-Cruz, C. H., R. E., Nuclear Instruments and Methods in
Downer, M. C.,and Maloney, P. J., Appl. Physics Research, B23,557,1987.
Phys. Lett., 53,654,1988.
29. Koren, G. and Yeh, J. T. C., J. Appl. Phys.,
11. Coche, M., Berthoud, T., Mauchien,P.,and 7,2120,1984.
Camus, P., Applied Spectroscopy, 43,646,
30. Deshmukh, S. and Rothe, E. W., J. Appl.
1989.
Phys., 66,1370,1989.
12. Iida, Y., Applied Spectroscopy, 43,229,1989.
31. Al-Wazzan, R. A., Lewis, C. L. S., and
13. Balazs, L., Gijbels, R., and Vertes, A., Ana- Morrow, T., Rev. Sci. Instrum., 67,85,
1996.
lytical Chemistry, 63,314,1991.
32. Al-Wazzan, R. A., Hendron, J. M., and
14. Mason, K. J. and Goldberg, J. M., Applied Morrow, T.,Applied Sulface Science, 96-98,
Spectroscopy, 45,1444,1991. 170,1996.
15. Owens, M. and Majidi, V., Applied Specto- 33. Bulatov, V., Xu, L., and Schechter, I., Ana-
scopy, 45,1463,1991. lytical Chemistry, 68,2966,1996.
16. Lee, Y., Sawan, S. P., Thiem, T. L., Teng, 34. Multari, R. A., Foster, L. E., Cremers,D. A.,
Y. Y., and Sneddon, J., Applied Spectros- and Ferris, M. J., Applied Spectroscopy, 50,
copy, 46,436,1992. 1483,1996.
17. Marine, W., Gerri, M., d'Aniello, J. M. S., 35. Multari, R. A. and Cremers, D. A., IEEE
Sentis, M., Delaporte, P., Forestier, B., and Transactions on P h m Science, 24,39,1996.
Fontaine, B., Applied Sulface Science, 54,
264,1992. 36. Nemet, B. and Kozma, L., Fresenius J. Anal.
Chem., 355,904,1996.
18. Kuzuya, M., Matsumoto, H., Takechi, H.,
and Mikami, O., Applied Spectroscopy, 47, 37. Kurniawan, H. and Kagawa, K., Applied
1659,1993. Specroscopy, 51,304,1997.
38. Martin, P., Campos, J., and Santander, 56. Hakkanen, H. J. and Korppi-Tommola,
J. L. G., J. Quant. Spectrosc. Radiat. Trans- J. E. I., Applied Spectroscopy, 49,1721, 1995.
fer, 57, 459, 1997.
57. Sabsabi, M. and Cielo, P., Applied Spectros-
39. Granse, G., S. Vollmar, Lenk, A., Rupp, copy, 49, 499, 1995.
A., and Rohr, K., Applied Sulface Science,
96-98, 97, 1996. 58. Ciucci, A., Palleschi, V., Rastelli, S.,
Barbini, R., Colao, F., Fantoni, R., Palucci,
40. Aguilera, J. A,, Aragon, C., and Campos, A., Ribezzo, S., and Steen, H. J. L. v. d., J.
J., Applied Spectroscopy, 46, 1382, 1992. Appl. Phys. B , 63, 185, 1996.
41. Hader, W., Tech mitt Kmpp Engl. Ed., 2,97, 59. E r s t , W. E., Farson, D. F., and Sames,
1992. D. J., Applied Spectroscopy, 50, 306, 1996.
42. Lorenzen, C. J., Carlhoff, C., Hahn, U., 60. Geertsen, C., Lacour, J. L., Mauchien, P.,
and Jogwich, M., J. Analytical Atomic Spec- and Pierrard, L., Spectrochimica Acta, 51B,
trometry, 7, 1029, 1992. 1403, 1996.
43. Sabsabi, M., Cielo, P. G., Body, S., and 6 1. Marquardt, B. J., Goode, S. R., and Angel,
Chaker, M., SPIE, 2069, 191, 1993. S. M., Analytical Chemistry, 68, 977, 1996.

44. Thiem, T. L., Salter, R. H., Gardner, J. A., 62. Miziolek, A. W., Optics and Photonics News,
Lee, Y . I., and Sneddon, J., Applied Spec- 39, 1996.
troscopy, 48, 58, 1994. 63. Radziemski, L. J., Loree, T. R., and
45. Thiem, T. L. and Wolf, P. J., Microchemi- Cremers, D. A., Optical Sciences, 39, 1983.
cal Journal,_SO, 244, 1994. 64. Palleschi, V., Arca, G., Ciucci, A., Rastelli,
46. Borthwick, I. S., Ledmgham, K. W. D., S., and Tognoni, E., IEEE-International
and Singhal, R. P., Spectrochimica Acta, 47B, Geoscience and Remote Sensing Symposium,
1259, 1992. 2, 854, 1996.
65. Vadillo, J. M. and Laserna, J. J., Talanta,
47. Allen, T. M., Kelly, P. B., Anderson, J. E.,
43, 1149, 1996.
Taylor, T. N., and Nogar, S. S., J. Appl.
Phys., 61, 221, 1995. 66. Vadillo, J. M., Palanco, S., Romero, M. D.,
and Laserna, J. J., Fresenius J. Anal. Chem.,
48. Anderson, D. R., McLeod, C. W., English, 355, 909, 1996.
T., and Smith, A. T., Applied Spectroscopy,
49, 691, 1995. 67. Kim, D. E., Yoo, K. J., Park, H. K., Oh,
K. J., and Kim, D. W., Applied Spectros-
49. Arnold, S. D. and Cremers, D. A., AIHA copy, 51, 22, 1997.
Journal, 56, 1180, 1995.
68. Maravelaki, P. V., Zafiropulos, V., Kiliko-
50. Sattmann, R., Sturm, V., and Noll, R., glou, V., Kalaitzaki, M., and Fotakis, C.,
J. Appl. Phys. D, 28, 2181, 1995. Spectromchimica Acta, 52B, 41, 1997.
51. Davies, C. M., Telle, H.H., and Williams, 69. Armstrong, R. L., J. Appl. Phys., 56, 2142,
A. W., Fresenius J. Anal. Chem., 355, 895, 1984.
1996.
70. Alexander, D. R. and Armstrong, J. G.,
52. Cremers, D. A., II, J. E. B., and Koskelo, Applied Optics, 26, 533, 1987.
A. C., Applied Spectroscopy, 49, 857, 1995.
71. Chylek, P., Jarzembski, A., and Chou,
53. Cremers, D. A., Applied Spectroscopy, 41, N . Y . , A p p l . Phys. Letr., 49, 1475, 1986.
572, 1987.
72. Hsieh, W. F., Eickmans, J. H., and Chang,
54. Bescos, B., Castano, J., and Urena, A. G., R. K., J. Optical Sociery of America, 4, 1816,
Laser Chemistry, 16, 75, 1995. 1987.
55. Gonzalez, A., Oritz, M., and Campos, J., 73. Biswas, A., Latif, H., and Radziemski, L. J.,
Applied Spectroscopy, 49, 1632, 1995. Applied Optics, 27, 2386, 1988.
74. Zheng, J., Hsieh, W. F., Chen, S., and 91. Arca, G., Ciucci, A., Palleschi, V., Rastelli,
Chang, R. K., Optics Letters, 13, 559, 1988. S., and Togononi, E., IEEE-International
Geoscience and Remote Sensing Symposium,
75. Hammer, D. X., Thomas, R. J., Noojin,
2, 856, 1996.
G. D., Rockwell, B. A., Kennedy,P. K., and
Roach, W. P., IEEE Journal of Quantum 92. Radziemski, L. J., Loree, T. R., Cremers,
Electronics, 32, 670, 1996. D. A., and Hoffman, N. M., Analytical Chem-
istry, 55, 1246, 1983.
76. Kitamori, T., Yokose, K., Suzuki, K.,
Sawada, T., and Gohshi, Y., Japanese Jour- 93. Eickmans, J. H., Hsieh, W. F., and Chang,
nal of Apllied Physics, 27, L983, 1988. R. K., Applied Optics, 26, 3721, 1987.
77. Vogel, A., Busch, S., and Parlitz, U., 94. Essien, M., Radziemski, L. J., and Sneddon,
J. Acoust. SOC.Am., 100, 148, 1996. J., J. Analytical Atomic Spectroscopy, 3,985,
1988.
78. Sacchi, C. A., J. 0pt.ical Society of America,
8, 337, 1991. 95. Archontaki, H. A. and Crouch, S. R.,
Applied Spectroscopy, 42, 741, 1988.
79. Pinnick, R. G., Biswas,A., Pendleton, J. D.,
and Armstrong, R. L., Applied Optics, 31 96. Ng, K. C., Ayala, N. L., Simeonsson, J. B.,
No. 3, 311, 1992. and Winefordner, J. D., Analytica Chimica
Acta, 269, 123, 1992.
80. Nyga, R. and Neu, W., Optics Letters, 18,
747, 1993. 97. Parigger, C. and Lewis, J. W. L., Applied
Physics Communications, 12, 163, 1993.
81. Feng, Q., Moloney, J. V., Newell, A. C.,
Wright, E. M., Cook, K., Kennedy, P. K., 98. Poulain, D. E. and Alexander, D. R., Ap-
Hammer, D. X., Rockwell, B. A., and plied Spectroscopy, 49, 569, 1995.
Thompson, C. R., IEEE Journal of Quantum
99. Haisch, C., Paine, U., and Niessner, R.,
Electronics, 33, 127, 1997.
Technical Digest- European Quantum Elec-
82. Cremers, D. A., Radziemski, L. J., and tronics Conference, 25, 1996.
Loree, T. R., Applied Spectroscopy, 38,721,
100. Kumar, V. and Thareja, R. K., J. Appl.
1984.
Phys., 64, 5269, 1988.
83. Wachter, J. R. and Cremers, D. A., Applied
101. Parigger, C., Plemmons, D. H., Hornkohl,
Spectroscopy, 41, 1042, 1987.
J. O., and Lewis, J. W. L., J. Quant.
84. Aragon, C., Aguilera, J. A., and Camp-, Spectrosc. Radiat. Transfer, 52, 707, 1994.
J., Applied Spectroscopy, 47, 606, 1993.
102. Yagi, T. and Huo, Y., Applied Optics, 35,
85. Stolarski, D. J., Hardman, J., Bramlette, 3183, 1996.
C. M., Noojin, G. D., Thomas, R. J.,
103. Yalcin,S., Crosley,D. R.,Smith, G. P., and
Rockwell, B. A., and Roach, W. P., SPIE,
Faris, G. W., Hazardous Waste and Hazard-
2391, 100, 1995.
ous Materials, 13, 51, 1996.
86. Ito, Y., Ueki, O., and Nakamura, S.,
104. Cremers, D. A. and Radziemski, L. J., Ana-
Analytica Chimica Acta, 299, 401, 1995.
lytical Chemistry, 55, 1252, 1983.
87. Nakamura, S., Ito, Y., Sone, K., Hiraga,
H., and Kaneko, K., Analytical Chemistry, 105. Cremers, D. A. and Radziemski, L. J., Ap-
68, 2981, 1996. plied Spectroscopy, 39, 57, 1985.

88. Knopp, R., Scherbaum, F. J., and Kim, 106. Sneddon, J., Trends in Analytical Chemistry,
J. I., Fresenius J. Anal. Chem., 355, 16, 1996. 7, 222, 1988.
89. Ho, W. F., Ng, C. W., and Cheung, N . H., 107. Ottesen, D. K., Wang, J. C. F., and
Applied Spectroscopy, 51, 87, 1997. Radziemski, L. J., Applied Spectroscopy, 43,
967, 1989.
90. Paksy, L., Nemet, B., Lengyel, A., Kozma,
L., and Czekkel, J., Spectrochimica Acta, 108. Morris, J. B., Forsch, B. E., and Miziolek,
51B, 279, 1996. A. W., Applied Spectroscopy, 44,1040,1990.

289
109. Cheng, E. A. P., Fraser, R. D., and Eden, 114. Lazzari, C., Rosa, M. D., Rastelli, S.,
J. G., Applied Spectroscopy, 45, 949, 1991. Ciucci, A., Palleschi, V., and Salvetti,
A., Laser and Particle Beams, 12, 525,
110. He, K. X., Hammond, T. D., Winstead,
1994.
C.B., Gole, J. L., and Dixon, D. A., J. Chem
Phys, 10, 7183, 1991. 115. Zhang, H., Singh, J. P., Yueh, F. Y., and
Cook, R. L., AppliedSpectroscopy,49,1617,
111. Joseph, M. R. and Majidi, V., J. Trace and
1995.
Microprobe Techniques, 10,207, 1992.
116. Nordstrom, R. J., Applied Spectroscopy, 49,
112. Casini, M., Harith, M. A., Palleschi, V.,
1490, 1995.
Salvetti, A., Singh, D. P., and Vaselli, M.,
Laser and Particle Beams, 9, 633, 1991. 117. Parigger, C., Plemmons, D. H., and Lewis,
J. W., Applied Optics, 34, 3325, 1995.
113. Flower, W. L., Peng, L. W., Bonin, M. P.,
French, N. B., Johnsen, H. A., Ottesen, 118. Haisch, C., Niessner, R., Matveev, O.,
D. K., Renzi, R. F., and Westbrook, L. V., Panne, U., and Omenetto, N., Fresenius
Fuel Processing Technology, 39, 217, 1994. J. Anal. Chem., 356, 21, 1996.

290