DTIC ADA519718: Comparison of Laser-Induced Breakdown Spectra of Organic Compounds with Irradiation at 1.5 and 1.064 micrometers

1.8 MB·English

by  Defense Technical Information Center

Comparison of laser-induced breakdown spectra
of organic compounds with irradiation
1:064 μm
at 1.5 and
Diane M. Wong and Paul J. Dagdigian*
Department of Chemistry, The Johns Hopkins University, Baltimore, Maryland 21218-2685
*Corresponding author: [email protected]
Received 17 June 2008; revised 29 August 2008; accepted 3 September 2008;
posted 4 September 2008 (Doc. ID 97536); published 9 October 2008
Acomprehensiveinvestigationoflaser-inducedbreakdownspectroscopy(LIBS)at1:500μmofresiduesof
sixorganiccompounds(anthracene,caffeine,glucose,1,3-dinitrobenzene,2,4-dinitrophenol,and2,4-dini-
trotoluene)onaluminumsubstratesispresentedandcomparedwithLIBSattheNd:YAGfundamental
wavelengthof1:064μm.Theoverallemissionintensitieswerefoundtobesmallerat1:500μmthanat
1:064μm,andtheratiosofC2 andCNmolecularemissionstotheHatomicemissionswereobserved
tobeless.PossiblereasonsfortheobserveddifferencesinLIBSat1:064μmversus1:500μmarediscussed.
©2008OpticalSocietyofAmerica
OCIScodes: 140.3440,300.6360,300.2140.
1. Introduction irradiation was compared with irradiation at longer
wavelengths. The authors state, without much de-
Laser-induced breakdown spectroscopy (LIBS) has
tail, that the energy required to generate a laser-
emerged as a powerful technique for the detection
induced plasma is lower at the longer wavelengths,
and characterization of materials such as residues
from1.064 to 1:470μm.They also report aprelimin-
of organic compounds, including explosives, on sur-
aryLIBSspectrumforlaserirradiationat1:470μm.
faces [1,2]. The overwhelming majority of modern
In the present study, we compare LIBS spectra of
LIBSexperimentsemployeithertheNd:YAGfunda-
organic residues on aluminum substrates obtained
mental beam at 1064nm or one of its
from plasmas generated at two separate irradiation
harmonics (532, 355, 266nm) as the laser source. wavelengths, 1.064 and 1:500μm. Since the laser
Thislaserisparticularlyconvenientforportablesen-
beamswereofthesameenergyandtransversebeam
sors because of its small size and power require- profile,therelativeintensitiesoftheatomicemission
ments. However, the fundamental wavelength lines and the overall signal strength for these two
(1064nm) has a very low threshold for damage to wavelengths could be directly compared. A set of or-
the eye as compared to slightly longer wavelengths ganic compounds with differing atomic molar ratios
of 1:5–1:6μm [3]. This 1:5–1:6μm spectral range is and chemical structure was investigated.
also advantageous because it lies within an atmo-
spheric“window,”forwhichthereislittleabsorption
2. Experimental
bymoleculespresentinair.Toourknowledge,there
AschematicdrawingoftheapparatusfortheseLIBS
isonlyonepublishedreportonLIBSwithlaserirra-
experiments is presented in Fig. 1. The pulsed laser
diation in this spectral region, namely, a brief study
byBaueretal.[4].Inthiswork,LIBSwith1:064μm radiation for generating the plasma was obtained
from the idler output of an optical parametric
oscillator (OPO) system (Continuum Panther). This
0003-6935/08/31G149-09$15.00/0 OPO was pumped by 290mJ of the 355nm tripled
©2008OpticalSocietyofAmerica output of an injection-seeded Nd:YAG laser system
1November2008/Vol.47,No.31/APPLIEDOPTICS G149
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
2008 2. REPORT TYPE 00-00-2008 to 00-00-2008
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Comparison of laser-induced breakdown spectra of organic compounds W911NF-06-1-0446
with irradiation at 1.5 and 1:064 μm
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT
Department of Chemistry,The Johns Hopkins NUMBER
; 50351.14
University,Baltimore,MD,21218-2685
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
U.S. Army Research Office, P.O. Box 12211, Research Triangle Park,
11. SPONSOR/MONITOR’S REPORT
NC, 27709-2211 NUMBER(S)
50351.14
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT

15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION 18. NUMBER 19a. NAME OF
OF ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 10
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
ICCD was synchronized with the laser pulse by the
output of a digital delay generator (Stanford Re-
search Systems), which was triggered with the out-
put of a photodiode that viewed the residual signal
output of the OPO.
The compounds were prepared as residues on
aluminum foil substrates. The compounds [anthra-
cene, caffeine, glucose, 1,3-dinitrobenzene (DNB),
2,4-dinitrotoluene (DNT), and 2,4-dinitrophenol
(DNP)] were obtained from Sigma-Aldrich, Fluka,
or Eastman and wereused without further purifica-
tion. The molecular formulas and atomic molar ra-
tios for the compounds investigated are given in
Table 1. For each compound, a 10–50μl aliquot of
Fig.1. Schematicdiagramoftheexperimentalapparatus.PDde-
a nearly saturated solution in methanol, acetone,
notesphotodiode.
or ethanol was delivered to the substrate surface,
and the solution was allowed to dry in a chemical
(Continuum Precision Powerlite 8000) operated at a fume hood. The resulting spot size was measured
10Hz repetition rate. The OPO output beam was to determine the average surface coverage of the
passed through three dichroic mirrors to remove compound. The surface concentrations ranged from
visiblesignalbeamandresidual355nmpumpbeam ∼20to140μg=cm2andwereapproximatelythesame
radiation from the near-IR idler beam. The OPO
for a given compound at the two wavelengths.
was tuned by rotating the BBO crystal in the
It is well known that emission on N and O atomic
oscillator cavity to yield either 1.500 or 1:064μm id-
lines in the red appear in laser-induced breakdown
ler radiation.
spectrainairwithnslasers,becauseofentrainment
Single pulses of idler radiation (typically 7mJ
of ambient air in the plasma [6,7]. Similarly, CN
in a 5mm diameter beam, with a wavelength-
emissionhasbeenobservedinLIBSofhydrocarbons
independentpulsewidthof∼6ns,asmeasuredwith
in ambient air [8]. Several approaches have been
a fast photodiode on the signal beam) were passed
taken to reduce or eliminate signals due to entrain-
through a mechanical shutter (Thor Labs) and fo-
cused with a fused silica lens (focal length 75mm) mentofairintoplasma.Double-pulseLIBShasbeen
shown to reduce significantly the effects of entrain-
onto an aluminum foil substrate coated with the
organic residue under investigation. The foil was ment [6]. As an alternative, observation of prompt
mounted on a motorized translation stage, and a emission with irradiation from a low laser fluence
freshareaoftheresiduewasirradiatedwitheachla- hasbeenemployedsincethissignalisrelativelyun-
sershot.Thepositionofthelensabovethesubstrate contaminated by air entrainment [7]. In this study,
was the same for the two wavelengths. It should be we have eliminated the contribution to N and O
notedthatthefocallengthofthelens,whichdepends atomic and CN molecular emission due to air en-
on n−1, where n is the refractive index [5], should trainment by bathing the substrate with argon
only differ by 1% between the two wavelengths. gas. This procedure allowed us to record signals on
The transverse intensity profile of the OPO laser C,H,N,andOlinesandtheCNbandscharacteristic
beam had an approximate top-hat shape, and it of the organic residue only. For bare aluminum sub-
wasnotpossibletopredictapriorithefocusedbeam strates,signalsduetoNandOarenotobserved.Be-
spot size. To get an estimate of the spot sizes at the low we discuss H and C signal levels from the bare
two wavelengths, areas of laser-induced craters on substrates. CN and C2 emission was observed only
bare aluminum foil were measured with an optical
with organic residues deposited on aluminum foil
microscopeequippedwithacamera.Thecraterareas
substrates, and not bare foil.
were essentially the same at the two wavelengths
and were approximately 0:040mm2 in magnitude.
Theopticalemissionfromtheplasmawasfocused
onto a UV-transmitting optical fiber (50μm core) Table1. MolecularFormulasandMolarRatiosfortheOrganic
withapairofoff-axisparabolicmirrors(focallength CompoundsInvestigated
25:4mm). The output of the optical fiber was trans-
MolarRatio
mitted to an echelle spectrometer (Andor Mechelle
5000) equipped with a gated, intensified CCD cam- Compound Formula C=H O=H N=H
era (Andor DH 734-18-03). For each compound, the Anthracene C14H10 0.714 0 0
ICCD delay and gate width were optimized for the Caffeine C8H10N4O2 0.800 0.200 0.400
best signal-to-background ratio, and the same set- Glucose C6H12O6 0.500 0.500 0
tings were employed for a given compound at both 1,3-Dinitrobenzene C6H4N2O4 1.500 1.000 0.500
wavelengths.TheICCDdelaysandgatewidthswere 2,4-Dinitrophenol C6H4N2O5 1.500 1.250 0.500
in the range of 1–3μs and 5–10μs, respectively. The 2,4-Dinitrotoluene C7H6N2O4 1.167 0.667 0.333
G150 APPLIEDOPTICS/Vol.47,No.31/1November2008
gest that there are significant differences in LIBS
spectra with irradiation at 1.500 versus 1:064μm.
For a quantitative comparison of LIBS spectra at
thetwowavelengths,twosetsof25single-shotspec-
tra were recorded for each compound at each wave-
length. For each spectrum, the intensities of the
transitions listed in Table 2 were measured. While
several sequences of the C2 d−a band system were
visible in the spectrum displayed in Fig. 2(a), we
measuredtheintensityofonlythestrongest,namely,
theΔv¼0sequence,sincetheC2emissionwasweak
or not detectable in some spectra. For the stronger
features, both peak areas and heights were mea-
sured, and their ratios were found to be consistent
to within ∼2%. For weaker features, notably the N
atomiclines,areascouldnotbereliablydetermined.
We hence employed peak heights as the measure of
intensity for all transitions. For CN and C2, the
heightsofthe(0,0)bandheadsweremeasured.Since
the N atomic lines were weak, the intensities of the
three lines (see Table 2) were summed in order to
haveamorereliableestimateoftheNatomicinten-
sity.Inallcases,intensitieswerecorrectedfor back-
ground continuum emission. No correction for the
wavelength-dependentresponseofthedetectionsys-
tem was carried out.
Fig. 2. Survey single-shot LIBS spectra of 2,4-DNT residues We inspected bare aluminum foil substrates for
(average surface concentration ∼75μg=cm2) on aluminum foil spectral features due to unintentional organic con-
forirradiationat(a)1.064and(b)1:500μm.Prominentemission
tamination. The foil substrates were carefully
featuresaremarked:Al,308and394nm;C,248nm;H,656nm;
handled with forceps to minimize the contact with
CN B−XΔv¼0 sequence, 388nm; C2, d−aΔv¼þ1, 0, −1 se-
quences at 474, 516, and 564nm, respectively. The lines in the fingers and to prevent the transfer of oils from the
700–800nmspectralrangearemainlyduetotheargonbathgas. fingertips onto the surface of the clean aluminum
foil.Insomecases,weakemissiononCandHatomic
lineswasobserved;however,thesesignalswereneg-
3. Results ligible compared to the intensities with an organic
In Fig. 2, we present examples of single-shot LIBS residueonthesubstrate.Wedidobservevaryingin-
spectra of 2.4-DNT irradiated at 1.064 and at tensitiesofboththeCandHlinesduetoorganicre-
1:500μm. Both spectra were taken under the same sidues with different commercial sources of
experimental conditions, except for the irradiation aluminum substrate, in particular with an alumi-
wavelength. The strongest lines in the spectra are num sheet of quoted purity 99.998% (metals basis)
at 394 and 308nm and can be identified as atomic bought from Alfa Aesar.
Al transitions originating from the aluminum foil Itiswell known thatLIBS intensitiesvarysignif-
substrate.Linesfromtheargonbathgascanbeseen icantly from shot to shot, as was the case in the
inthe700–800nmspectralrange.TheCandHatom- presentexperiment.Ithasbeenfoundthatmeasure-
ic emission lines from the DNT residue, seen near ment of ratios of intensities yields more robust
248 and 656nm, respectively, are clearly visible in results [9–13]. We hence determined ratios of inten-
thespectra.NandOatomiclinesintheredarepre- sitiesineachsingle-shotspectrum.Theseratioswere
sent in the spectra but are obscured by the stronger then averaged for each set of 25 single-shot spectra.
Arlines.Severalmolecularemissionfeaturesareob-
servableinthespectra,includingtheCNB−XΔv¼
0sequencenear 388nm and the Δv¼þ1, 0, and −1 Table2. SpectralTransitionsUsedforQuantitativeAnalysisof
sequencesoftheC2 d−abandsystem(Swanbands) LIBSSpectraofOrganicCompounds
near516nm.Theoff-diagonalCNB−XΔv¼þ1and Species Wavelength(nm) Transition
−1sequenceshavesmallFranck–Condonfactorsand H 656.3 n¼3→n¼2
are hence barely visible in the spectra. C 247.9 2p3s1P1→2p21D2
It can be seen that the CN and C2 bands are par- O 777.2–5 3p5P1;2;3→3s5S2
ticularlystrongintheDNTspectrumforirradiation N 742.4 3p4S3=2→3s4P1=2
at1:064μm,displayedinFig.2(a).Bycontrast,these N 744.2 3p4S3=2→3s4P3=2
bands are much weaker for irradiation at 1:500μm N 746.8 3p4S3=2→3s4P5=2
[see Fig. 2(b)]. The differences in the intensities of C2 516.3 d−aΔv¼0
CN 388.3 B−XΔv¼0
features in the two spectra displayed in Fig. 2 sug-
1November2008/Vol.47,No.31/APPLIEDOPTICS G151
Nospectra were discarded incomputing average in- atomiclineintensityforbothdatasetsofeachofthe
tensityratios.SincetheHlinewasusuallythestron- organic compounds investigated at the two irradia-
gest atomic feature originating from the organic tion wavelengths. Similarly, Fig. 4 presents the
residues, ratios of intensities with respect to the H measured ratios of molecular emission intensities,
atomic line were computed. involving C2 and CN, to the H atomic line intensity.
Figure3presentsthemeasuredratiosoftheC,O, WeseefromFigs.3and4that,withafewexceptions,
andNatomiclineintensities(peakheights)totheH there is good consistency between the ratios
Fig.3. RatiosofC,O,andNlineintensities(peakheights)totheHatomiclineintensityforLIBSoforganicresiduesonaluminumfoil
substrates.Thespectrawererecordedunderidenticalconditionsexceptfortheirradiationwavelengths,whichareindicatedatthetopof
thesetsofpanels.Thetwoplottedpointsforeachcompoundrepresentmeansoftheintensityratiosfortwosetsof25single-shotspectra;
theerrorbarsarethestandarddeviationsofthemean.
G152 APPLIEDOPTICS/Vol.47,No.31/1November2008
determined for the two data sets of each compound themolarC=Hratiosofthecompoundsinvestigated
and wavelength. (compare Fig. 3 with Table 1). However, such a cor-
The dinitro compounds displayed the largest C=H relation does not apply to the O=H and N=H ratios.
intensityratiosforbothirradiationwavelengths,but The most dramatic differences in intensity ratios
the ratios were not dramatically greater than those between the two wavelengths are seen in the
of the other compounds. There were only slight dif- ratios of the molecular (CN and C2) features to H
ferencesintheC=Hintensityratiosforagivencom- atomicintensity(seeFig.4).WeobserveC2 emission
pound between the two wavelengths. The O=H only for the aromatic organic compounds. Strong C2
intensity ratios at a given wavelength were similar emission has previously been reported for LIBS of
for the compounds investigated. For all the com- this class of compounds and been found to increase
pounds, the O=H intensity ratios were larger for ir- with increasing numbers of fused aromatic rings
radiation at 1:500μm than at 1:064μm, but again [8,14]. DNT has by far the largest C2=H intensity
the differences between the two wavelengths were ratio (∼16) for irradiation at 1:064μm. In marked
not great. At 1:064μm, DNP and DNT displayed contrast, the C2=H intensity ratios for all the aro-
the largest N=H intensity ratios; these ratios were matic compounds at 1:500μm are much smaller, ap-
significantly smaller at 1:500μm. Caffeine and proachingunity.Infact,DNThasthesmallestC2=H
DNBhadsimilar,smallN=Hintensityratiosatboth intensity ratio at this wavelength.
wavelengths. Similar, large differences are observed for the
We observe that the measured C=H intensities at ratios of CN to H emission for irradiation at
both irradiation wavelengths roughly correlate with 1.064 versus 1:500μm. For LIBS at 1:064μm, the
Fig.4. RatiosofC2 andCN(0,0)bandheadintensitiestotheHatomiclineintensityforLIBSoforganicresiduesonaluminumfoil
substrates. The spectra were recorded under identical conditions except for the irradiation wavelengths, which are indicated at the
topofthesetsofpanels.Thetwoplottedpointsforeachcompoundrepresentmeansoftheintensityratiosfortwosetsof25single-shot
spectra;theerrorbarsarethestandarddeviationsofthemean.
1November2008/Vol.47,No.31/APPLIEDOPTICS G153
nitrogen-containing aromatic compounds display wavelengthoftheline.Theradiativetransitionprob-
CN=H intensity ratios ranging from ∼6 to 15. The abilities A were taken from the NIST Atomic Spec-
i
corresponding intensity ratios are seen in Fig. 4 tra Database [16].
to be much smaller at 1:500μm. The only other Temperatures were determined in this way for
nitrogen-containingmolecule,caffeine,showedsmall LIBS of residues of two of the organic compounds,
CN=H ratios at both wavelengths. namely, anthracene and DNT. These are presented
It is also of interest to compare the overall inten- in Table 3. The reported temperatures, which were
sities at the two irradiation wavelengths. Of the obtained from intensities measured with fairly wide
atomiclines,theHatomiclineisthemostappropri- detector gates (see Table 3), are similar to those re-
atetoconsidersincetheintensityratiosdisplayedin portedbyPiehleretal.[15]fordelaytimesof∼10μs.
Figs.3and4arereferencedtothisline.Figure5dis- In this experiment, somewhat higher laser powers
plays the absolute H line intensities for both data were employed (35mJ of 1:064μm radiation focused
sets of each of the organic compounds investigated witha50mmfocallengthlens).Ofmostinterestfor
at the two irradiation wavelengths. Since the C2 the interpretation of our measured LIBS intensities
and CN emissions are quite strong for the aromatic isthecomparisonofthetemperaturesforthetwoir-
compounds, the intensities of the bands due to the radiation wavelengths. We see from Table 3 that for
species are also displayed in the lower panels both compounds the temperature deduced for irra-
of Fig. 5. diation at 1:500μm is noticeably higher than for ir-
WeseefromFig.5thattheHatomiclineintensity radiation at 1:064μm.
isthestrongestforanthracene,caffeine,andglucose Anestimateoftheelectrondensitycanbeobtained
at both wavelengths, while the intensity of this line fromtheobservedbroadeningoftheHatomiclineat
is less for the nitroaromatic compounds. Wealso ob- 656:3nm.Therelationshipbetweenelectrondensity
serve in Fig. 5 that the H line intensity is approxi- andlinewidthiswellestablishedformanyelements
mately a factor of 2 smaller at 1:500μm than [17], and we have employed the tables prepared by
at 1:064μm for anthracene, caffeine, and glucose, Vidal et al. [18]. Figure 6 compares the profile of
while the H line intensities for the nitroaromatic theHatomic656:3nmlineforirradiationofanthra-
compounds are only slightly different at the two cene and DNT residues at the two wavelengths. It
wavelengths. can be seen that for both compounds the profiles
Significant differences are found for the absolute forirradiation at 1:500μm are slightly broader than
intensities of the molecular emissions between the for1:064μm.Thiswasalsotrueforthislinewiththe
two wavelengths. For all the aromatic compounds, other compounds investigated. From comparison of
the C2 and CN emission is significantly smaller at the line profiles displayed in Fig. 6 and the profiles
1:500μm than at 1:064μm. At both wavelengths, computed by Vidal et al. [18], we estimate that the
the C2 emission is stronger for anthracene than electron density, averaged over the detection gate
for the nitroaromatic compounds. We also observe (seeTable3),was6×1016 and8×1016cm−3 forirra-
thattheCNemissionisweakeratbothwavelengths diationofanthraceneat1.064and1:500μm,respec-
for caffeine than the nitroaromatic compounds. tively.Thecorresponding electrondensitiesforDNT
Some comments about the intensity of Al atomic were 3×1016 and 4×1016cm−3. We hence find
transitions, which arise from ablation of the alumi- slightly higher electron densities for irradiation at
numfoilsubstrate,areinorder.Wedoobservesome 1:500μm than at 1:064μm.
variationintheintensitiesofthe394and308nmAl
atomic transitions for the different organic residues
4. Discussion
at a given irradiation wavelength. However, there
This study is to our knowledge the first comprehen-
were no dramatic differences in the intensities of
sivestudyofLIBSat1:5μm.Itisperhapsnotunex-
these transitions between the two wavelengths.
pected that we were able to obtain LIBS spectra at
For the interpretation of these results, it is useful
this irradiation wavelength since laser radiation of
to estimate the temperature and electron density of
sufficientlyhighintensity atanywavelengthshould
the plasmas generated at the two irradiation wave-
cause ablation, and also formation of plasma, at a
lengths. We have utilized the ratio of the intensities
of the Al atomic lines at 394.4 and 308:2nm for the surface. Of greater importance is the quantitative
comparison between LIBS at 1:500μm and at the
determination of temperature, similar to the deter-
Nd:YAGfundamentalwavelengthof1:064μm.From
minationscarriedoutbyPiehleretal.[15]inastudy
the absolute intensities displayed in Fig. 5 and the
ofLIBSofaluminuminvariousbathgases.Theratio
intensity ratios displayed in Figs. 3 and 4, we see
oftheintensitiesIofthetwolinescanbeemployedto
thattheoverallLIBSintensityissignificantlysmal-
estimatetheplasmatemperaturewiththefollowing
lerat1:500μmthanat1:064μmfortheinvestigated
equation:
(cid:1) (cid:2) organic residues on aluminum substrates. We also
I1 ¼g1A1λ2exp −E1−E2 ; ð1Þ findsignificantlydifferentratiosofintensitiesofvar-
I2 g2A2λ1 kT ious emission features in LIBS spectra at the two
wavelengths. The most dramatic differences are in
Here,gi andEi arethestatisticalweightandenergy, the ratios of the intensities of the C2 and CN mole-
respectively, of the upper level of line i and λ is the cularbandsrelativetotheHatomiclineintensityfor
i
G154 APPLIEDOPTICS/Vol.47,No.31/1November2008
the aromatic compounds. Moreover, the absolute in- 1:064μm could be due to differences in a variety of
tensitiesofthesebandsaresmallerat1:500μmthan parameters, including the amount of material ab-
at 1:064μm. lated, plasma temperature, and electron density of
The weaker emission intensities for irradiation at the laser-induced plasma. Amoruso et al. [19] have
1:500μm as compared to those for irradiation at presented a comprehensive review of laser-ablation
Fig.5. IntensitiesoftheHatomiclineandtheC2andCN(0,0)bandheadsforLIBSoforganicresiduesonaluminumfoilsubstrates.The
spectrawererecordedunderidenticalconditionsexceptfortheirradiationwavelengths,whichareindicatedatthetopofthesetsofpanels.
Thetwoplottedpointsforeachcompoundrepresentmeansoftheintensityratiosfortwosetsof25single-shotspectra;theerrorbarsare
thestandarddeviationsofthemean.
1November2008/Vol.47,No.31/APPLIEDOPTICS G155
plasmas and have discussed ablation and formation
ofaplasmabyirradiationwithpulsedlaserradiation
ofnsduration.Theleadingedgeofthelaserpulseis
absorbed by the substrate, and there is sufficient
time for thermal diffusion into a zone larger than
thefocusedlaserspotsize.Theorganicresiduespre-
pared in this study were thin enough that the laser
energyisabsorbedpredominantlybytheconduction
electrons in the underlying aluminum substrate.
Thisenergyistransferredtothelattice,andneutral
and ionized species evaporate from the surface. The
ns laser pulse is sufficiently long that the incipient
plasma absorbs radiation from the tail of the laser
pulse by the inverse bremsstrahlung process [20],
heatingtheplasmaandshieldingthesubstratefrom
further ablation.
The absorption of laser energy by the aluminum
substrate should not differ greatly between the two
laserwavelengths(1.064versus1:500μm)employed
in this study. However, inverse bremsstrahlung
absorption by the plasma will be greater at the
longer wavelength [20]. This additional energy up-
takebytheplasmaperhapsprovidesanexplanation
for our higher observed temperatures and electron Fig.6. ProfilesoftheHatomiclineat656:3nm,averagedover25
densities at 1:500μm (see Table 3 and Fig. 6). This single-shotspectra,forLIBSat1:064μm(solidlines)and1:500μm
would appear to be inconsistent with the fact that (dottedlines)ofanthraceneandDNTresiduesonaluminumfoil
substrates.
emission intensities were found to be lower at
1:500μm than at 1:064μm, since an increased tem-
perature and electron density would be expected to
tions. It would be interesting to follow the time de-
lead to higher emission intensities, all other para-
pendence of the molecular emissions with a
meters remaining unchanged. It may be that less
material is ablated from the sample at 1:500μm be- narrower detector gate, but this is beyond the scope
causeoftheincreasedshieldingbytheplasmaatthis of this investigation.
wavelength.
Thefinalobservationtobediscussedisthesignif-
5. Conclusion
icantlyreducedmolecularemissionforirradiationat
1:500μm as compared to emission at 1:064μm. The A comprehensive investigation of LIBS at 1:500μm
molecularspeciesC2andCNcanarisefromfragmen- oforganicresiduesonaluminumsubstrateshasbeen
tation of the parent compound or recombination of presented and compared with LIBS at the Nd:YAG
the atoms in the plasma [7,14,21]. Baudelet et al. fundamentalwavelength1:064μm.Theoverallemis-
[7] have noted that it should be possible to distin- sionintensitieswerefoundtobesmallerat1:500μm
guishemissionfromthesetwomechanismsbyvary- thanat1:064μm,andtheratiosofmoleculartoatom-
ing the detector gate delay. With the relatively long ic emissions were observed to be less. This result
delays employed in the present study, the observed could have implications for the detection sensitivity
molecularemissionisprobablyduetoatomicrecom- of LIBS at 1:500μm. It would be desirable to carry
bination. The reduced molecular emission at out a similar comparison of LIBS at these wave-
1:500μm could be the result of lower atomic concen- lengths with higher-power laser radiation. This will
trations from the increased plasma shielding dis- require a different laser system as the maximum
cussed above, since the molecular concentration pulseenergyat1:500μmoftheOPOemployedinthis
will depend nonlinearly on the atomic concentra- study is ∼7mJ.
Theauthorsgreatlyacknowledgetheadvicegiven
Table3. TemperaturesDeterminedfromRelativeIntensitiesof to us by the LIBS group at the Army Research La-
theAlAtomic394.40and308:22nmLinesforLIBSPlasmas boratory(A.W.Miziolek,K.L.McNesby,F.C.DeLu-
IrradiationWavelength cia, Jr., J. L. Gottfried, and C. A. Munson) at
Aberdeen Proving Ground, Maryland, on the acqui-
Compound 1:064μm 1:500μm
sitionofLIBSspectra.Theloanofamotorizedtrans-
Anthracenea 3580(cid:1)80 3980(cid:1)80 lation stage by J. B. Spicer is also gratefully
2,4-Dinitrotolueneb 3160(cid:1)50 3940(cid:1)60 appreciated. This research has been supported in
aDetectorgatedelayandwidth1.25and7μs,respectively. part by the U.S. Army Research Office under grant
bDetectorgatedelayandwidth1.25and10μs,respectively. W911NF-06-1-0206.
G156 APPLIEDOPTICS/Vol.47,No.31/1November2008
References breakdown spectroscopy,” Spectrochim. Acta Part B 57,
1131–1140(2002).
1. A. W. Miziolek, V. Palleschi, and I. Schechter, eds., Laser-
12. F.Ferioli,P.V.Puzinauskas,andS.G.Buckley,“Laser-induced
Induced Breakdown Spectroscopy (LIBS): Fundamentals
breakdownspectroscopyforon-lineengineequivalenceratio
andApplications(CambridgeU.Press,2006).
measurements,”Appl.Spectrosc.57,1183–1189(2003).
2. D. A. Cremers and L. J. Radziemski, Handbook of Laser-
13. S.Kaski,H.Häkkänen,andJ.Korppi-Tommola,“Determina-
InducedBreakdownSpectroscopy(Wiley,2006).
tionofCl=CandBr=Cratiosinpureorganicsolidsusinglaser-
3. AmericanNationalStandardsInstitute,AmericanStandard
induced plasma spectroscopy in near vacuum ultraviolet,”
forSafeUseofLasers(LaserInstituteofAmerica,2000).
J.Anal.At.Spectrom.19,474–478(2004).
4. C. Bauer, P. Geiser, J. Burgmeier, G. Holl, and W. Schade,
“Pulse laser surface fragmentation and mid-infrared laser 14. L. St.-Onge, R. Sing, S. Béchard, and M. Sabsabi, “Carbon
spectroscopyforremotedetectionofexplosives,”Appl.Phys. emissionfollowing1:064μmlaserablationofgraphiteandor-
B85,251–256(2006). ganicsamplesinambientair,”Appl.Phys.A69,S913–S916
(1999).
5. D. E. Gray, ed., American Institute of Physics Handbook
15. T.N.Piehler,F.C.DeLuciaJr.,C.A.Munson,B.E.Homan,A.
(McGraw-Hill,1972).
W.Miziolek,andK.L.McNesby,“Temporalevolutionofthe
6. F. C. DeLucia, Jr., J. L. Gottfried, C. A. Munson, and
A.W.Miziolek,“Double-pulselaser-inducedbreakdownspec- laser-inducedbreakdownspectroscopyspectrumofaluminum
metal in different bath gases,” Appl. Opt. 44, 3654–3660
troscopyofexplosives:Initialstudytowardsimproveddiscri-
mination,”Spectrochim.ActaPartB62,1399–1404(2007). (2005).
16. NISTAtomicSpectraDatabase,version3NISTAtomicSpec-
7. M.Baudelet,M.Boueri,J.Yu,S.S.Mao,V.Piscitelli,X.Mao,
and R. E. Russo, “Time-resolved ultraviolet laser-induced traDatabase,version3.http://physics.nist.gov/PhysRefData/
breakdownspectroscopyfororganicmaterialanalysis,”Spec- ASD.
trochim.ActaPartB62,1329–1334(2007). 17. H. R. Griem, Spectral Line Broadening by Plasmas
8. A. Portnov, S. Rosenwaks, and I. Bar, “Emission following (Academic,1974).
18. C. R. Vidal, J. Cooper, and E. W. Smith, “Hydrogen Stark-
laser-inducedbreakdownspectroscopyoforganiccompounds
inambientair,”Appl.Opt.42,2835–2842(2003). broadening tables,” Astrophys. J. Suppl. Ser. 25, 37–136
(1973).
9. R. Sattmann, I. Mönch, H. Krause, R. Noll, S. Couris,
19. S.Amoruso,R.Bruzzese,N.Spinelli,andR.Velotta,“Charac-
A. Hatziapostolou, A. Mavrromanolakis, C. Fotakis,
E. Larrauri, and R. Miguel, “Laser-indeuced breakdown terization of laser-ablation plasmas,” J. Phys. B 32, R131–
spectroscopy for polymer identification,” Appl. Spectrosc. R172(1999).
52,456–461(1998). 20. J. F. Ready, Effects of High Power Laser Radiation (Aca-
10. M.Tran,Q.Sun,B.W.Smith,andJ.D.Winefordner,“Deter- demic,1971).
minationof C∶H∶O∶N ratios in solid organic compounds in 21. V. I. Babushok, F. C. DeLucia, Jr., P. J. Dagdigian,
laser-induced plasma spectroscopy,” J. Anal. At. Spectrom. J.L.Gottfried,C.A.Munson,M.J.Nusca,andA.W.Miziolek,
16,628–632(2001). “Kineticmodelingstudyofthelaser-inducedplasmaplumeof
11. L.St.-Onge,E.Kwong,M.Sabsabi,andE.B.Vadas,“Quan- theexplosivecyclomethylenenitramine(RDX),”Spectrochim.
ActaPartB62,1321–1328(2007).
titativeanalysisofpharmaceuticalproductsbylaser-induced
1November2008/Vol.47,No.31/APPLIEDOPTICS G157