DTIC ADA519477: Thermodynamic and Spectroscopic Properties of Nd:YAG-CO2 Double-Pulse Laser-Induced Iron Plasmas

0.6 MB·English

by  Defense Technical Information Center

SpectrochimicaActaPartB64(2009)961–967
ContentslistsavailableatScienceDirect
Spectrochimica Acta Part B
journal homepage: www.elsevier.com/locate/sab
Thermodynamic and spectroscopic properties of Nd:YAG–CO Double-Pulse
2
☆
Laser-Induced Iron Plasmas
⁎
Matthew Weidman, Santiago Palanco, Matthieu Baudelet, Martin C. Richardson
TownesLaserInstitute,CREOL–TheCollegeofOptics&Photonics,UniversityofCentralFlorida,P.O.Box162700,Orlando,FL32816-2700,UnitedStates
a r t i c l e i n f o a b s t r a c t
Articlehistory: Double-PulseLaser-InducedBreakdownSpectroscopyofironusingbothNd:YAGandTEA–CO lasershas
2
Received15December2008 beeninvestigatedtobetterunderstandmechanismsofsignalenhancement.Thesignaldependenceonthe
Accepted27July2009 delaybetweenthetwolaserpulsesshowsanenhancedsignalwhentheCO laserpulseinteractswiththe
2
Availableonline4August2009
samplebeforetheNd:YAGpulse.SignalkineticsandasimplemodelofsampleheatingbytheCO pulseshow
2
that the enhancement during the first 700ns is due primarily to sample heating. Images of the sample
Keywords:
surfaceafterablationaswellastime-integratedpicturesoftheplasmasuggestthatparticlesareejectedfrom
Laser-inducedbreakdownspectroscopy
the surface during the first microseconds after the arrival of the CO pulse and provide fuel for the
DoublepulseLIBS 2
Laserablation subsequentplasmacreatedbytheNd:YAGlaser.
Laserheating ©2009ElsevierB.V.Allrightsreserved.
Laser-inducedparticles
1.Introduction (anglebetween thelaserbeams), differentlaserwavelengths and/or
pulsedurationsandinterpulsedelay,aswellasdifferentpulseenergies.
Optical emission spectroscopy of laser-induced plasmas is a Acollinearapproachcombinestwopulsesfollowingthesamepath
powerful spectroscopic diagnostic technique for sensing and other and focused upon the same point of the sample. The orthogonal
applicationsbecauseofitsabilitytoidentifyalltypesofmaterials(solid, configurationusesmultipleopticstofocusatdifferentlocations(double-
liquid,gas),ataclosedistanceaswellasinastand-offconfiguration. focus collinear configuration has also been demonstrated [2]). Two
Despitetheincreasingpopularityofthistechnique(alsocalledLIBS— configurationsarepossiblewithanorthogonalgeometry:apre-ablation
Laser-Induced Breakdown Spectroscopy) its sensitivity and precision spark to produce plasma above the sample prior to the ablation or a
are relatively poor compared to other well established analytical reheatingsparkfocusedupontheexpandingplasmacreatedbythefirst
techniques.Thisisbecauseoftheuncertaintyinlaserenergycoupling pulse.
tothesample,significantmatrixeffectsandrelativelyhighbackground From femtosecond to microsecond pulses, several combinations
signal in the atomic and ionic spectra, among other reasons. One have been studied: fs–fs [3], fs–ns [4], ns–ns [5], ns–µs [6]. Other
approachtoimprovingsensitivity(byincreasingthesignal-to-contin- works have studied the influence of the combination of different
uum-backgroundratio)istheuseofadouble-pulseconfiguration(DP– wavelengthsfromultraviolet(UV)tonear-andmid-infrared(NIRand
LIBS).TheaimofDP–LIBSistoimprovethecouplingofthelaserenergy MIR):UV–NIR[5],Visible–NIR[8],NIR–NIR[5],NIR–MIR[6],NIR–MW
tothetargetandtotheablatedmaterial,leadingtoahighernumberof [9]andthecombinationofpulseswithdifferentenergy.
emittersfromtheanalyzedsampleintheplasma. In Babushok et al. [10] and Scaffidi et al. [4], overviews of the
Since the first demonstration of double-pulse LIBS in 1969 by different mechanisms leading to emission enhancement in DP–LIBS
PiepmeierandMalmstadt[1],ithasbeendemonstratedthattwolaser experiments are given. Depending on the delay time between the
pulses for LIBS lead to enhanced emission intensities, longer plasma pulses,severalexplanationstotheenhancementaregiven.Fromzero-
lifetimes and higher plasma temperatures. Tooptimize theseeffects, delaytotensofms,thephysicsbehindthephenomenonchanges.In
severalcombinationshavebeenproposedinvolvingvariousgeometries Scaffidi et al., three mechanisms are mainly detailed: Pulse-plasma
coupling(re-excitingorexcitingoftheplasmacomponents),sample
heating (heating by the hot plasma close to the surface resulting in
☆ Thispaperwaspresentedatthe5thInternationalConferenceonLaser-Induced increasedablation)andatmosphericeffects(creationofahotregion
Breakdown Spectroscopy (LIBS 2008), held in Berlin, Adlershof, Germany, 22–26 withlowdensityabovethesample).InBabushoketal.,anextensivelist
September2008,andispublishedintheSpecialIssueofSpectrochimicaActaPartB, ofpossibleeffectsisgivenfromthepreparationofthesamplebythefirst
dedicatedtothatconference.
⁎ Correspondingauthor. pulse(smoothing,firstcrater,onsetofmelt)totheenergycouplingwith
E-mailaddress:[email protected](M.C.Richardson). theplasma(ionyield,plasmare-heating,electrondensity).
0584-8547/$–seefrontmatter©2009ElsevierB.V.Allrightsreserved.
doi:10.1016/j.sab.2009.07.023
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
2009 2. REPORT TYPE 00-00-2009 to 00-00-2009
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Thermodynamic and spectroscopic properties of Nd:YAG?CO2 W911NF-06-1-0446
Double-Pulse Laser-Induced Iron Plasmas
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
Townes Laser Institute, CREOL ? The College of Optics & NUMBER
; 50351.17
Photonics,University of Central Florida,P.O. Box
162,Orlando,FL,32816-2700
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)
50351.17
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 7
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
962 M.Weidmanetal./SpectrochimicaActaPartB64(2009)961–967
In this paper, we present the results of a double-pulse LIBS symmetricresonatorandemitsat10.6µm.Anintra-cavityirisisused
experimentwhereananosecondNIRpulseiscombinedwithaMIR tolimitthenumberofhigherordertransversemodesandprovidea
CO laserpulseandfocusedonanironsample.An8-foldenhancement circularlysymmetricbeamprofileasviewedusinganIRpyroelectric
2
isreachedbyirradiatingthesamplefirstwiththeCO laserpulseand beamcamera(SpiriconPyrocamIII).Thelaserenergyiscontinuously
2
thenablatingitwiththeNd:YAGlaser.Theeffectofinterpulsedelay tunablebetween70μJand80mJbyusingasetofcrossedpolarizers.
allows us to advance a combination of mechanisms leading to the Thepulseenergyismeasuredusingthesamepyroelectricjoulemeter
enhancement of the LIBS signal compared to that obtained with a described above. The peak-to-peak and RMS fluctuations in pulse
singleNd:YAGlaserpulse. energy are 3.5% and 0.98%, respectively. The temporal profile is
measured with a multiple junction photovoltaic detector (Vigo
2.Experimentalsetup Systems PVM-10.6) with nanosecond response time. The temporal
profileiscomposedofa100nspulsefollowedbya1μstail(Fig.7a).
TheexperimentalconfigurationisshowninFig.1.Twolasersare Thelaserisfocusedwitha26cmfocallengthlensat13degreesfrom
fired onto the same spot of a sample and the light emitted by the normal,andthelens-to-sampledistanceis25.5cm.Thebeamradius
plasmaiscollectedandguidedtoaspectrometer.Thespatialoverlap atthesamplelocationisapproximately280μmasdeterminedusinga
ofthetwopulsesisshownontheinsetinFig.1. knife-edgescan.Anirradianceof30×106W/cm2(assumingapulse
durationof1µs)wasobtainedwiththelaseroperatingat75mJper
2.1.Lasersources pulse.
TheNd:YAGlaser(QuantelBrilliant)isQ-switchedandoperatedat 2.2.Collectionopticsandspectrometer
thefundamentalwavelengthof1064nm.Itisexternallytriggeredat
0.33Hz as limited by the CO laser. A thin-film polarizer is used in Theplasma emission iscollectedusingan f/1 quartz lensand is
2
conjunctionwithaλ/2platetoprovidecontinuouscontrolofthelaser coupledtoa1mround-to-lineUV-gradefiberbundleusingasecond
energy from 5mJ up to 340mJ without changing the flashlamp (f/2)lens.Thecollectionopticsareat40°fromthesurfacenormal.The
voltageortheQ-switchdelaytime.Thepulseenergyismeasuredwith fiber bundle is f-matched to a 0.5m Czerny-Turner imaging
a calibrated pyroelectric joulemeter (Gentec SOLO 2 with a QE-25 spectrometer (Acton 2500i, Princeton Instruments) with a 1800
head). The peak-to-peak and RMS fluctuations in pulse energy are lines/mm grating. The light is detected using an intensified charge
1.51%and0.44%respectively.Thefullwidthhalfmaximum(FWHM) coupled device (Andor iStar DH720-25F-03) with 1024×256pixels
pulsedurationis5nsasmeasuredwithafastphotodiode(Thorlabs) anda25mmGenIIintensifier.Theintensifierdelayismeasuredfrom
placedbehindadielectricturningmirror.Thelaserisfocusedwitha therisingedgeofthesecondpulse.Eachspectrumistheaverageof30
13.5cmfocallengthlensatnormalincidence,andthelens-to-sample singleacquisitions.
distanceis13cm.Thebeamradiusatthesamplelocation,determined
bytheknife-edgescantechnique,isapproximately230µm.TheNd: 2.3.Timing
YAG laser was operated at 39mJ per pulse, corresponding to an
irradianceof4.7×109W/cm2. Thetimingoftheexperimentiscontrolledusingtwodigitaldelay
TheCO laserisatransverselyexcitedatmospheric(TEA)CO laser generators(StanfordResearchModelsDG-645andDG-535).TheDG-
2 2
withcomponentsmadebyLumonics.Itutilizesanexternal1mhalf- 645triggerstheflashlampoftheNd:YAGlaserandalsotriggersthe
DG-535at10Hz.TheDG-535dividesthefrequencyto0.33Hzand
triggerstheQ-switchoftheNd:YAGlaser,theCO laserandtheICCD
2
camera. Timing uncertainty between Nd:YAG and CO pulses is
2
primarily limited by the jitter in the CO laser pulse onset and has
2
peak-to-peak fluctuations of approximately 42ns and a standard
deviation of 9ns. The timing configuration for the interpulse delay
andtheacquisitionisshowninFig.2.
2.4.Samples
Thetargetsampleswerea100μm-thickfoilof99.9+%pureiron
(Product#:356808,Sigma-Aldrich,St.Louis,MO,USA).Eachfoilwas
cleaned with acetone and mounted so that it remained flat and
verticaltominimizeanydepositionofmaterialaftertheablation.The
Fig.2.Timingconfigurationoftheexperiment.TheinterpulsedelayTpissetaspositive
whentheNd:YAGlaserprecedestheCO2laserpulseandnegativeintheoppositeorder.
The acquisition delay Td is set to 1µs after the second pulse and the duration of
Fig.1.Experimentalsetup. acquisitionTwto25ns.
M.Weidmanetal./SpectrochimicaActaPartB64(2009)961–967 963
samplewastranslatedbetweeneachdoublepulselaserirradiationin timewaschosenfordetermininginstantaneousplasmatemperature.
ordertopresentafreshlocationforeachshot. Spectraforthesingle-pulsecaseofNd:YAGandCO areshownina)
2
andb)respectively,andthedouble-pulsecaseforNd:YAGpreceding
3.Resultsanddiscussion theCO pulseby1μsandNd:YAGfollowingtheCO pulseby1μsare
2 2
showninc)andd)respectively.Inallcases,theenergyofNd:YAGand
Theoriginalintentionofthisdouble-pulseLIBSstudywastouse CO laserswere23Jcm−2and30Jcm−2respectively.
2
10.6µm radiation from the CO laser to re-heat or raise the ThespectrallinesinFig.3aareofsingly-ionizediron.Theweak
2
temperatureoftheplasmacreatedbytheNd:YAGlaserafterithad signalinFig.3bcanbeattributedtominimalablatedmassandlow
diffusedawayfromthesurface.Sincethecriticaldensityofaplasma, plasmatemperatureduringtherelativelylonginteractiontimeofthe
the point at which light is reflected, scales as the square of the CO pulse. For the double pulse case, with the CO laser pulse
2 2
wavelength, reheatingtheplasma(Fig.3c),thereisnosignalenhancementrelative
(cid:1) (cid:3) to the signal acquired from the Nd:YAG only. This can be partially
Ncr = 2πc 2ε0me attributed to the acquisition delay time of 1μs measured from the
e e λ2 second laser pulse; therefore, 2µs after the Nd:YAG pulse. This
measurementdoesnotimplysignificantplasmaheatingbytheCO
2
longerwavelengthlightwillinteractwiththeplasmafartherdown pulse.However,theeffectsofplasmareheatingusingaCO laserpulse
2
thedensitygradientinthecoronaoftheplasmaproducedbytheNd: (increasedlineintensities)havebeendemonstratedpreviously.
YAG laser. With the relatively long scale lengths produced in the InworkbyKillingeretal.[6],anenhancementofbothatomic(up
plasma,inverseBremsstrahlungabsorption(IBA)occursatdensities toafactorof60)andionic(uptoafactorof300)aluminumlinesfora
wellbelowthecriticaldensity. highpurityaluminasamplewasobservedwhenaNd:YAGpulse(5ns,
Previousstudiesusingmultiplewavelengthsfordouble-pulseLIBS 50mJmm−2)wasfollowedbyaCO pulse(100nspeakand5μstail,
2
hasbeenreported[5]usingbothUVandNIRpulses,andacorrelation 70mJmm−2).Thedifferenceinbehaviorcomparedtoourresultscan
wasfoundbetweensignalenhancementandplasmatemperature[5] be attributed to both the different energy densities used, since the
showingthatthesignalincreaseswiththetemperature. signalfromananosecondlaserplasmaisapproximatelyproportional
Acomparisonbetweensingle-pulseanddouble-pulseLIBSspectra toI3/2[7],Ibeingthelaserintensity,andsampleproperties.Ametal
is shown in Fig. 3. Each spectrum represents the average of 30 samplewasusedinourcaseandaninsulatorintheircase,eachhaving
separateshotsobtainedwith1μsdelay(measuredwithrespecttothe differentthermodynamicproperties.Acompletecomparisoncannot
secondlaserpulse)andanacquisitiontimeof25ns.Thisacquisition be made with our work since the timing configuration is not
mentionedintheirpublication.
TounderstandtheeffectoftheCO pulseindouble-pulseLIBS,the
2
time delay between pulses has been varied from the CO pulse
2
precedingtheNd:YAGpulseby5mstotheCO pulsefollowingNd:
2
YAGpulseby10μs.Forclarity,Fig.4showstheintensityoftheFe(II)
274.65nm ionic line for a delay between −10μs and +10μs. A
negativedelaymeanstheCO pulsereachesthesamplebeforetheNd:
2
YAGpulse.AlthoughthereisnoplasmareheatingbytheCO pulse,a
2
signalenhancementofapproximately8timesoccursforaninterpulse
delay of −1μs. This signal enhancement is closely correlated to an
enhancementinplasmatemperature,ascalculatedusingaBoltzmann
plot[11]withtheparameterslistedinTable1.Thetemperaturefora
Fig.3.LIBSspectraofironfroma)asingleNd:YAGpulse,b)asingleCO2laserpulse, Fig.4.Background-correctedpeakintensityoftheFeII274.65nmlineasafunctionof
c)doublepulsespectrawithCO2followingNd:YAGby1µscomparedwithNd:YAG theinterpulsedelay.AnegativedelaymeanstheCO2pulsereachesthesamplesbefore
aloneLIBSspectrumandd)doublepulsespectrawithCO2precedingNd:YAGby1µs. theNd:YAGpulse.Apositivedelayisfortheoppositetimingconfiguration.Comparison
(Acquisitiondelay:1µs,durationofacquisition:25ns).Ina),thedottedlineshowsa ismadewiththesingle-pulsecases.Theerrorbarsrepresentthestandarddeviation
background spline interpolation and the stars indicate the lines used for plasma over30singleacquisitions.Theinsetintheupperrightcornerisazoombetween−3µs
temperaturecalculation.Baselinecorrectionisappliedforallspectra. and0µsshowingthemaximumenhancementat−1µs.
964 M.Weidmanetal./SpectrochimicaActaPartB64(2009)961–967
Table1 given in Table2. Thetemporalprofile of theheat sourceI (t) is an
0
Parametersofthespectrallinesusedintheplasmatemperaturecalculations[12]. experimental profile acquired from a 75mJ CO pulse. The model
2
Wavelengthλul UpperlevelenergyEu Degeneracy TransitionprobabilityAul yields an estimate of the temperature on the sample surface and
(nm) (eV) gu (s−1) doesn'tincludeplasmashieldingorablationofthesamplebythelaser
268.48 8.43 10 1.4108 (theenergyatthesurfaceisconsideredcompletelytransformedinto
269.26 8.37 12 1.2108 heat).Thelackofabsorptionofthelaserpulsebytheplasmaleadsusto
270.40 7.97 8 1.2108 consider that this simple model will give a higher estimate of the
271.44 5.55 6 5.5107 temperatureatthesurfaceofthesample.However,noconclusionson
273.96 5.51 8 1.9108
theablationproductsortheplasmacharacteristicscanbeextracted.
274.65 5.59 6 1.9108
274.93 5.55 8 2.1108 Thecalculatedtemperaturedistributioninthetargetisplottedasa
275.57 5.49 10 2.1108 function of time as shown in Fig. 5. The first conclusion is that the
278.37 7.70 10 7.0107 meltingpointT (1811Kforiron)isnotreachedwiththeirradianceof
m
theCO laserpulseused.Themaximumtemperatureatthesurfaceis
2
approximately 860K and occurs at 550ns. This is supported by the
single-pulseNd:YAGplasmais15000±2000Kand23000±2000K microscope images of the sample irradiated by only the CO laser
2
fordoublepulsewith−1µsinterpulsedelay. (Fig.6b)wheresmalldegradationofthematerialbytheCO pulseis
2
WhentheCO pulseisfiredbeforetheNd:YAG,twomechanisms observedafter30shotsonthesamespot.Thecraterdepthafter30shots
2
mayoccur:(i)theCO pulseheatsthesamplebeforethearrivalofthe (using a Zygo NewView 6300 3D optical profiler) is not discernible
2
Nd:YAGpulse,(ii)theNd:YAGinteractswiththeweakplasmacreated withintheroughnessofthesamplebytheinstrumentandissmaller
by the CO pulse, ablating the sample softly and extracting non thantheabsorptionlength(30nm)whichisindicativeofnanometric
2
excitedparticlesfromthesamplewithsizedistributionrangingfrom layersbeingablatedbyeachCO pulse.Givenminimalmaterialremoval
2
atomictoclusters. by melting and vaporization, the Nd:YAG pulse following the CO
2
ablationisnotinfluencedbysignificantsampledeformation.
3.1.HeatingofthesamplebytheCO pulse Giventhemaximumtemperatureoccursatt=550ns,amaximum
2
signalenhancementataninterpulsedelayof1µsisnotcompletely
Several studies have shown a positive correlation between the explainedbysampletemperatureeffects.
intensity of the LIBS signal intensity and the sample temperature
[13,14].Thisphenomenonisexplainedbylessenergybeingrequired 3.2.InteractionoftheNd:YAGlaserpulsewithCO laserinducedplasma
2
tomeltorvaporizethesample—theimportantparametersinlong-
pulseablationaboveroomtemperature.Furthermore,thereflectivity WithlittleapparentsampledegradationasaresultoftheCO laser,
2
of metals decreases when the surface temperature increases [15]. acontinuum-likeemissionisdetectedasshowninFig.3b.
Thesetwofactorscouldleadtoabetterenergycouplingbetweenthe This continuum can be attributed to bremsstrahlung and/or poly-
Nd:YAGpulseandthesampleifpreviouslyheatedbytheCO pulse.To atomicclusters.TheseemittersareextractedbytheCO laserpulseduring
2 2
understandtheimportanceofthisphenomenon,asimplemodelof the interaction with the surface, when a nanometric layer is ablated.
heatconductionisused. Whenthelaserremovesmaterialfromthesurface,thesizedistributionof
Themodelisone-dimensional(1D)sincetheabsorptionlengthof particlesislimitedbytheabsorptiondepth1/α[17].Twotypesofemitters
a radiation with a 10.6µm-wavelength at 298K in the target is arelocalizedintime-integratedpicturesoftheplasmainthevisibleregion
approximately 30nm (lα=1/α, with α=3.3 107m−1), which is shown in Fig. 7b (using a manually triggered Nikon D40x DSLR with
muchsmallerthaneitherlaserbeamdiameter.The1Dheatequation 300mm1:1macrozoomlensatf/8with1to2secondexposuretimeto
hasthefollowingform: integratetheentireplasmalifetime).Lightemissionisobservedattwo
differentlocations:onecomponentalongthenormalandtheotherclose
" ! #
tothesurface.Theparaxialemissioncanbeattributedtosmallemitters
∂Tðx;tÞ ∂ κ ∂Tðx;tÞ α
= + Sðx;tÞ (atomsandions),ejectedwithhighspeedfromthesample(tensof
∂t ∂x C ρ ∂x C ρ
p p
withS(x,t)=I (t)e−ax(1−R)·I (t)isthelaserirradiance(inWcm−2)
0 0
andtheotherparametersarelistedinTable2.Thiscalculationisdone
via a matricial implicit algorithm for solving parabolic partial
differential equations[16]. Theconditionsofthe algorithmare also
Table2
Physicalandnumericalparametersusedinour1Dheatequationmodel.
Parameters Values
Thermalconductivity,κ(Wcm−1K−1) 0.802
Specificheat,Cp(Jg−1K−1) 0.449
Massdensity,ρ(gcm−3) 7.87
Absorptioncoefficient,α(m−1)(λ=10.6µm) 3.3107
Reflectivity,R(λ=10.6µm) 0.967
Meltingpoint,Tm(K) 1811
Spacerange [x0=0µm;xf=10µm]
Spacesampling 701points
Timerange [t0=−650ns;tf=1849ns]
Timesampling 2500points
Initialcondition T(x,t=0)=300K
Boundaryconditionsonx0andxf ∂Tðx∂=xx0;tÞ=0,∂Tðx∂=xxf;tÞ=0 Fsaigm.p5l.eCaalfcteurlattheeditrermadpiaetriaotnurweidthisttrhibeultaisoenripnutlhseetdaerpgiectteasdainfuFnigct.i7oan.o(fsetiempehfyosricaanliarnond
Thevaluecorrespondingtothesolidformisgivenbytheliterature[19]. algorithmicparametersinTable1).
M.Weidmanetal./SpectrochimicaActaPartB64(2009)961–967 965
Fig.6.Microscopepicturesoftheironsampleaftera)asingleNd:YAGlasershot,b)anaccumulationof30CO2lasershots,c)acombinationofaNd:YAGlasershotfollowedbyaCO2
lasershot1μslaterandd)acombinationofaCO2lasershotfollowedbyaNd:YAGlasershot1μslater.
kms−1)[18],whereasthemarginalemissioncanbeduetoslowheavy 3.3.CombinationofmechanismsforLIBSsignalenhancement
emitters (clusters) (hundreds of m s−1) [18]. This spatially divided
emissioniscomparedwiththelightemissionoftheplasmacreatedby MaximumsignalenhancementoccurswhentheCO precedesthe
2
theNd:YAGlaseronly(acquiredinthesameconditions)wheretheplasma Nd:YAGpulseby1µs.ThisconditionisshowninFig.8alongwiththe
ishomogeneouslyconfinedtothesurfaceanditsexpansioniscomparable normalizedplasmaemissionandcalculatedsurfacetemperature.We
tothebeamwaist. candividethetimescaleintwostages:
The spatially integrated temporal trace of the visible emission is
showninFig.8c.Thefirstpeakisduetobremsstrahlungfromelectronic 3.3.1.Thefirst700ns
collisionswithfastemitters,andisshownbytheparaxialemissionin Duringthefirst700ns,thereisastrongcorrelationbetweenthe
Fig. 7b. During the course of the CO laser pulse, there is a second surfacetemperatureandtheplasmaemission.The100nspeakofthe
2
emissionincreasepeakingat800ns.Thisisattributedtoheavyparticles CO laserpulsequicklyheatsthesamplefromapproximately300Kto
2
extracted from the surface during the relatively long CO pulse 650K.Thetailofthepulsereachesahotsurfaceandheatsthesample
2
evolution.Thelowvelocityoftheseemittersrestrictstheirpositionto to 860K. The signal follows the same trend until approximately
withinhundredsofmicronsofthesample,asseeninFig.7b. 700ns. This enhances the production of emitting products. These
When the Nd:YAG pulse arrives 1µs after the CO pulse, a products are mainly heavy emitters rather than ions or atoms
2
maximumcouplingwiththeCO inducedplasmaoccursbecausethe commontomostLIBSplasmasandthereislittlesurfacemodification,
2
maximum number of particles has been extractedfrom the sample leadingtotheconclusionthattheseheavyparticleshavenanometer
(since there is no excitation by the CO laser anymore). Particle size. When the Nd:YAG pulse reaches a hot surface, less energy is
2
ablationbytheNd:YAGlaserisefficientbecauseoftheirhighsurface- needed for ablation resulting in increased mass ablation and LIBS
to-mass ratio, and at this time an enhanced ionic LIBS spectrum is signal.
observed(Fig.3d)comparetothesurfaceablationbytheNd:YAGonly
(Fig.3a).Thereisasignificantdifferenceinthevisibleemissionforthe 3.3.2.Between700nsand10µs
case of CO pulse preceding the Nd:YAG pulse by 1μs and the CO The extraction of particles from the surface increases until
2 2
pulsefollowingtheNd:YAGpulseby1µsasreflectedinFig.7candd. approximately 0.7 to 1µs. At this time, the coupling between the
Theseimagesshowalessintensesurfaceplasmainthesecondcase CO induced plasma and the Nd:YAG pulse is maximum. Emission
2
suggesting the absorption of the Nd:YAG above the surface. This is fromtheablationofparticleswithadistributionofvelocityaround
alsoobservedinthecraterimagesshowninFig.6.Theedgesofthe 500mpersecondwillremainwithinthefieldofview(approximately
craterinducedbytheNd:YAGinthedoublepulsecaseofCO firstare 2mm)forapproximately4µs.However,noticeableenhancementof
2
smootherthaninthecasewhenasingleNd:YAGpulsereachesthe theLIBSsignaloccursuntil10µs.Furtherexperimentsareinprogress
samplefirst(Fig.6aandc). tobetterexplainthemechanismsofsuchalongenhancement.
966 M.Weidmanetal./SpectrochimicaActaPartB64(2009)961–967
Fig.7.Time-integratedpicturesoftheplasmaproducedbya)asingleNd:YAGlasershot,b)asingleCO2lasershot,c)acombinationofanNd:YAGlasershotfollowedbyaCO2laser
shot1μslaterandd)acombinationofaCO2lasershotfollowedbyaNd:YAGlasershot1μslater.Thearrowsrepresentthedirectionoftheincominglaserandthedottedlinesthe
laserbeamdiameter2w0.
4.Conclusion sample before the arrival of the Nd:YAG laser pulse during the first
700ns, without melting the sample or creating noticeable surface
Theopticalemissionfromlaser-inducedplasmainadouble-pulse deformation.Nevertheless,thebrightcontinuumemissionrepresenta-
LIBSexperiment(Nd:YAGat1064nmandCO laserat10.6µm)oniron tiveofheavyandslowmovingparticlesextractedandejectedduringthe
2
hasbeenstudiedtounderstandwhichmechanismsareresponsiblefor CO laserinteractionwiththeironsample.Theseparticlesactasfuelfor
2
signalenhancementinthisconfiguration.TheCO laserpulseheatsthe the second pulse, and lead to greatest signal enhancement approxi-
2
matelyonemicrosecondafterthebeginningoftheCO laserpulse.
2
Acknowledgments
ThisworkissupportedbytheUSARO-MURIprogramon“Ultrafast
LaserInteractionProcessesforLIBSandotherSensingTechnologies”,
andbytheStateofFlorida.
References
[1] E.H.Piepmeier,H.V.Malmstadt,Q-Switchedlaserenergyabsorptionintheplume
ofanaluminumalloy,Anal.Chem.41(1969)700–707.
[2] J.BruceIII,AlexanderDennis,Collineardualpulse-dualfocusenhancementof
femtosecond LIBS, 5th International Conference on Laser-Induced Breakdown
Spectroscopy(Berlin,Germany,2008),2008,22–26September.
[3] V.Pinon,C.Fotakis,G.Nicolas,D.Anglos,Doublepulselaser-inducedbreakdown
spectroscopywithfemtosecondlaserpulses,Spectrochim.ActaPartB63(2008)
1006–1010.
[4] J.Scaffidi,S.M.Angel,D.A.Cremers,Emissionenhancementmechanismsindual-pulse
LIBS,Anal.Chem.78(1969)24–32.
[5] L.St-Onge,V.Detalle,M.Sabsabi,Enhancedlaser-inducedbreakdownspectros-
copyusingthecombinationoffourth-harmonicandfundamentalNd:YAGlaser
pulses,Spectrochim.ActaPartB57(2002)121–135.
Fig.8.OptimaltimingforthemaximumofLIBSsignalenhancement:a)CO2laserpulse [6] D.K.Killinger,S.D.Allen,R.D.Waterbury,C.Stefano,E.L.Dottery,Enhancementof
intensity,b)Nd:YAGlaserintensity,c)theplasmaemissionfromtheCO2pulseonlyand Nd:YAGLIBSemissionofaremotetargetusingasimultaneousCO2laserpulse,
d)thecalculatedsurfacetemperatureofthesample. Opt.Express15(2007)12905–12915.
M.Weidmanetal./SpectrochimicaActaPartB64(2009)961–967 967
[7] S.Palanco,C.López-Moreno,J.J.Laserna,Design,constructionandassessmentofa [13] S.Palanco,L.M.Cabalín,D.Romero,J.J.Laserna,Infraredlaserablationandatomic
field-deployable laser-induced breakdown spectrometer for remote elemental emissionspectrometryofstainlesssteelathightemperatures,J.Anal.At.Spectrom.
sensing,SpectrochimicaActaPartB61(2006)88–95. 14(1999)1883–1887.
[8] C.Gautier,P.Fichet,D.Menut,J.-L.Lacour,D.L'Hermite,J.Dubessy,Studyofthe [14] S.H. Tavassoli, A. Gragossian, Effect of sample temperature on laser-induced
double-pulse setup with an orthogonal beam geometry for laser-induced breakdownspectroscopy,Opt.LaserTechnol.41(2009)481–485.
breakdownspectroscopy,Spectrochim.ActaPartB59(2004)975–986. [15] M. von Allmen, A. Blatter, Laser-beam interactions with materials — physical
[9] B.Kearton,Y.Mattley,Laser-inducedBreakdownspectroscopy—sparkingnew principlesandapplications,2ndedition,Springer,Berlin,1998.
applications,Nat.Photonics2(2008)537–540. [16] J.P.Nougier,Méthodesdecalculnumérique,3rdedition,Masson,Paris,1985.
[10] V.I.Babushok,F.C.DeLuciaJr.,J.L.Gottfried,C.A.Munson,A.W.Miziolek,Double [17] E.N.Sobol,Phasetransformationsandablationinlaser-treatedsolids,JohnWiley
pulselaserablationandplasma:laserinducedbreakdownspectroscopysignal &Sons,NewYork,1995.
enhancement,Spectrochim.ActaPartB61(2006)999–1014. [18] S.Noel,J.Hermann,T.E.Itina,Nanoparticlegenerationinplasmasproducedby
[11] D.A.Cremers,L.J.Radziemski,Handbookoflaser-inducedbreakdownspectroscopy, ultra-shortlaserpulses,Appl.Surf.Sci.253(2007)6310–6315.
JohnWiley&SonsLtd,Chichester,2006. [19] D.R.Lide,Handbookofchemistryandphysics,82ndedition,CRCPress,NewYork,
[12] Y.Ralchenko,A.E.Kramida,J.ReaderandNISTASDTeam(2008).NISTAtomicSpectra 2001.
Database (version 3.1.5), [Online]. Available: http://physics.nist.gov/asd3 [2008,
December2].NationalInstituteofStandardsandTechnology,Gaithersburg,MD.