1064nmNd：YAG激光诱导铁等离子体特征参数的研究

本资料包含pdf文件1个，下载需要1积分

第 25卷第 7期2013年 7月强 激 光 与 粒 子HIGH POW ER LASER AND PARTICLE束BEAM SVoI.25，No.7Ju1.，2013Article ID： 1001-4322(2013)07-1690-07M easurements of iron plasma parameters producedby a 1 064 am pulsed Nd：YAG laserLuo W enfeng ， Zhao Xiaoxia ， Zhu Haiyan1， I i Dongdong ， I i Xiaoli(1·s 。。f。厂E c r。札 c Engi e盯 g，Xi口n Uni Prs t 。，Poss n d Tefec。mmn cn 。”s，x ，Ⅱ 710121，c n：2·s 。。 。，P s s n 彳Pcn Ⅱf& Kgectron E gi Pe ng， Xinn u er 。，Ars nnd sc Pnc ，xi，nn 71o065， nn)Abstract： A 1064 nli1 pulsed Nd：YAG laser is used for the ablation of an iron bar samp1e in ah at atmos-pheric pressure and the laserindueed plasma charaeteristics are examined.The electron number density 8. 7lu cm m the Iron Plasma is inferred from the Stark broadened profile of Fe I 376.553 nm averaged with lOsingle spectra. In orderto minimize relative err0rs in calculation of the electron temperature，an improved itera-tlve Boltzmann Pl叭 method is uged· Experimental results show that the electron temperature is 8O58 K with aregress on coefficient of 0·981 38· Based on the experimental results，the plasma is verified to be in 10cal therm。dynam c equI1ibrium (LTE)and free from self-absorption. Considering the laser phot0n frequency(2.82×1014 Hz)is l g than the plasma freqHency(8. 3×10 Hz)，the 1oss of energy due t0 reflection ofrhe jaserbeam trom he plasma can be assumed to be insignificant. Experiments also demonstrate that the inverseB emsstrahlung(IB)absorpti0n is the dominant photon absorptionProcess during the laser-plasma interaction。

and the corresponding IB absorption coeficient is 0. 14 cm 。

KeY words： atomic em ssion spectroscopy； laser- induced breakdown spectrosc0py； plasma； electrontemperature； electron number densitvCLC 叫 mber： 0432·12； TH744. 1 Document code：A doi：10. 3788/HPLPB201325O7.1 69Oaser-nducedbreakd0wn spectroseopy(LIBS)has rapidly developed into a major anaIytica1 techno1ogvw。th the capability of detecting all chemical elements in a sample， because of its rea1-time response， no n-con -tact opt caI nature and unnecessity of special sample preparation[ -引. In LIBS，a high-power 1aser Du1se is fo- cused onto a sample and the plasma or laser spark will be created when the breakdown threshold is reached。

ollectedby a lens or fiber optics， emissions from atoms and ions in the plasma are analyzed by a spectro-graph · l hese atomic or ionic emission lines allow for a qualitative identification of the species present inthe plasma，while their relat ve intensities can be used for quantitative analysisE引. Because the laser-inducedp asma s used as VaPor zat on atomization， excitation and ionization source， LIBS has been considered asone oi the most convement analytical methods for elemental analysis in manv fields ]。

Exper"nents haVe demonstrated that the interaction of the focused laser radiation with the target is acompIex phenomenon· rhe plasma characteristics depend not only on the laser properties(wavelength，pu1seduratlon and puIse energy)， but also on the physical and chemical properties of the target mate 1 as we11 asthe surroundmg atmospher c conditions. Although a lot of studies have investigated the parameters thatntluence the plasma characteristics，there is no model to completely describe the laser ablation phenomenonand study of its basic mechanisms is still a great challenge。

n cn s paPer， measurements of plasma temperature and electron density in iron plasma are reDorted。

w nlch。s±ormed by rrad ation of a solid iron bar in air at atmospheric pressure with a 1064 nm pulsed Nd：Y AG I se ·Knowledge of these two Variables is vita1 to predicting the dissociation，at。mizati。n，ionization。

F。 d ti。it m：Supported by the Y.oung scientists Fund。f the Nati。na1 Natural science Foundati。n。f China(611O8O61)， the Xi'an Uni-r1er、s，i ty o f P o sts、and Telecomm unicati0ns Foundati。n f0r Y。ung Teachers，Xian science and Techn ologPlan ning6P- 11roIoilelictt cx 89w、Lo and ScentIfIc Research Program Funded by Shaanxi Provincial Education DweplaArutlmuNeynt(2012013JK0620、 、 - B.0graphY：IJ 。w。feng(1 974- )，rlale，PhD，engaged in laser induced hreakd。wn spectr。sc。py；luowenfeng###xupt.edu.cn。

第 7期 Luo Wenfeng，et al：Measurements of iron plasma parameters produced by a 1064 nm pulsed 1691and excitation processes occurring in the plasma.At the same time，a quantitative method is provided for theselection of spectral lines to determine plasma temperature.Using this iterative Boltzmann plot algorithm ，the plasma temperature is determined with 3 3 iron atomic 1ines.whereas the electron number density is esti～mated from the Stark broadened profile of Fe I 376.553 nm.Based on the experimental results，the condi～tions of I TE and optically thin plasma are verified，and some other plasma parameters(plasma frequency，inverse Bremsstrahlung coefficient，ere.)are calculated as wel1。

1 Experim ental setupA schematic diagram of the experimental set-up isshown in Fig.1.The LIBS set-up uses a Nd：YAG laserdelivering 135 mJ per pulse at 1064 nm，with a pulseduration of 19.7 ns FW HM and a frequency of 1 Hz(SGR，Beamtech Optronics). The laser beam is fo-cused onto the sample using a quartz lens(厂- 150mm)，which is put on an XYZ translation stage so thatevery laser pulse is incident at a fresh location on thesample surface and thus crater formation is avoided。

One end (200，um in diameter)of a fiber bundle is usedto collect the light emitted from the Iaser-induced plas- Fig.1 LIBS experimental set-upma at the right angle to the direction of the plasma expansion，and its other five ends are connected to the en-trance slits(10 m in diameter)of the broadband spectrometer AvanSpec-2048FT-5 (200-720 nm range，0.06-0.08 optical resolution with 2400-1800 grooves/mm，Avantes，Holland)，which is triggered by theswitch of the Nd：YAG laser. In order to reduce statistical error due to 1aser shot-to-shot fluctuation.theoutput data without specific statement are averaged from ten laser shots and stored in a personal computerthrough AvaSoft-LIBS software for subsequent analysis. All the experiments are performed at room temper～ature in air and at atmospheric pressure。

2 Results and discussion2.1 Description of plasma emissionFig.2 shows segments of the typical emission spectra recorded from laser-induced iron Dlasma. Theplasma spectrum consists of a number of neutral as well as ionic characteristic emission lines. However。inthe spectra recorded at early plasma time，the signal is mostly dominated by the continuum emission which isattributed to the Bremsstrahlung process and recombination of electrons with ions。，. As time progresses。

continuum emission diminishes，while ionic and atomic emission lines become dominant. To obtain a goodcompromise between the signal background ratio(SBR)and the line emission intensity，a delay time(5 us)between the laser shot and the initiation of data acquisition is chosen in our experiments. At the same time，the integrated time(2 ms)is adopted.The assignment of these atomic and ionic lines shown bv the arrows inFig.2 is done using NIST databaseE. 2.2 Determination of excitation temperatureThe key parameters of laser-ablated plasma are density and temperature. The intensitv of line emissionfrom a plasma is influenced by the electron temperature and the number of atoms in the Dlasma which is re- lated to the level of ablation. It is essential to have the knowledge of the electron temperature and electrondensity in order to understand processes occurring in the laser-ablated plasmaE13]. Furthermore，Dlasma de-nptlons start by trying to characterize properties of the assembly of ions，electrons，molecu1es and atomsather than the individual species. Under LTE conditions，plasma properties can be described through theconcePt of temperature，and the population of the excited levels for each species follows the BoItzmann distri1692 强 激 光 与 粒 子 束 第 25卷800三600400-量 200-O 01200800400O200 240 280wavelength/nrn32O- ， - - - - ] I420 440 460 480 ，500 520wavelength/nm1000800砖 ； 600. 嚣 400甚200O础、 1200go0400O- - - - - - - - - - 20 340 360 380 400 420wavelength/nm500 540 580wavelength/nm n 620Fig 2 Segments of emission spectra of laser ind' d iron plasbution and the r re1ative spectral line intensity Im is giVen by[加]- - 象十h[ ] ㈩where ，A ，g ，h，c，E ，T ，是，己，(丁)and N (T)are the waVe1ength，transition probability，。t t tca1 weight of the upper level，Plank constant，speed of light in vacuum ，uppe level。n。 gy，。I。 t 。temper-ature，Boltzmann constant，partition function and tota1 number density of the species，respectively· Pl。ttmgthe exDressi。n。n the left-hand side。f the equati。n versus E， yields a slope of-1/(kTe)·Th。pl m 。m-perature can be obtained eve1 without knowing N(T)and U(T)。

energy/cmboenergy/cm.s Boltzmann plot with 141 Fe I- ol1 l - thmFi g.4 B olt z mann p lo t w ith 33 F e I mis sion。

1fitting correlation coefficient 0.266 27 tt g cor 。 The uncertainty in the temperature determination based on Eq·(1)c。mpris。 b。 h h。errOrs ere -tive intensity measurement and the usualy much more important unce nties of the trans t on probabili-ties[5].T。minimize the uncertainty in temperature determinati。n，spectra1 lines should be h。 as.tar。

- part as p。ssib1e in excitati。n energY.At the same time，the use Of a ser s Of nes fr。m di er 。x cltatlontates。 the same species ca lead t。greater prec 。n Othe plasma tempeI ure d mm 旧 · Order u ase d iron山 uⅢre ， an-improved i terati，ve Bol-tzm annpl ot me hod isuse d[i4]based。n the simp1e 1inear regress on.F tl4laFe I lines are。sef 1ec9t2ea73asKpiost。e lblttlaaitn，e...aditwu Ji. utha a poor re- Boltzmann plot shown in Fig.3.W ith the fitting result， temperature 0 上 厶 。工 g目 寸H第 7期 Luo Wenfeng，et a1：Measurements of iron plasma parameters produced by a 1064 nin pulse d. 1693gression coefficient R of 0.2 7.Second，the distances ofeach data to the straight line are calculated. ThesDectral line with the maximum distance is discardedfor it does not match with the overall trend，and then- the regression function is recalculated with the remai-ning spectral lines.This procedure is repeated until theregression coefficient reaches a threshold value of 0.98。

After 108 iterations，a regression coefficient of 0.98138 is reached and the plasma temperature of 8058 K isinferred from the slope shown in Fig.4 and their corre- Fiiteration number Variations of correlation coefficient a.d plasma temperattlresponding parameters of the 3 3 Fe I lines can he found in s iterati0n number for Fe I emission linesNIST databaseE.At the same time，Fig.5 also de-Dicts the variations of R and T。with the iteration number. W ith the outlier emission lines discarded，the re-gression coefficient increases and at the same time the plasma temperature becomes more and more stabie·The error of each datum is not more than 10% and is not plotted in the figures。

2.3 Determination of electron densityE1ectron density is an important parameter to establish plasma thermodynamic equilibrium and to under-stand the processes governing laser ablation.The electron density in plasma can be determined from spectralbroadening of an atomic line. There are three likely broadening mechanisms occurring in the laser-ablatedD1asma，i.e.Doppler broadening，resonance broadening and Stark broadening[ .The Doppler broadening，resulting frOm the motion of atoms，depends on the absolute temperature and the atomic mass of the emit-ting species[. △ D-7.2× 10- (T /M ) o (2)where M is the atomic mass of the element，and 。is the central wavelength of the spectral line.Take Fe I376.553 nm for example，the Doppler width is 0.003 nm with the electron temperature obtained above，which is too sma11 to be observable at the spectrometer resolution used in our experiments. Resonance broad-ening is a1so neg1ected because atomic state of Fe I 376.553 nm is not related to the resonance state. TheStark effect is caused by electrons in the plasma that perturb the energy levels of the individual atoms，whichin turn broadens the emission lines.The FW HM 1/2 of the Stark broadening is related to the electron densityby the expressionc 。]2∞(器)3.5c(器 (1-1.2 ，3) ( ) (3)where reDresents electron impact parameter，C is the ion broadening parameter，N is electron density，andNn is the number of particles in the Debye sphere.Considering that the perturbations caused by ions is negli-gible compared to electrons，Eq.(3)reduces to△ ll/2- 2 ( ) (4)Fig.6 shows the typical Stark broadened line profile of Fe I 376.553 nm averaged with 10 single spectrain order to smOoth out the fluctualions of laser-ablated plasma intensity and it is fitted fairly well with a typi-ca1 Lorentzian profile(R ：0.95).W ith the fitting results，the electron density is inferred to be 8.7× 1O”cm DT]. 2.4 LTE requirementTo determine the electron temperature，the plasma must satisfy the equilibrium conditions，i.e. theplasma must hold a state of LTE during the observation window.In an LTE plasma，the collisional excita-tion and de-excitation process must dominate radiative processes and this requires a minimum electron densi-1694 强 激 光 与 粒 子 束 第 25卷ty. The following criterion must be satisfied by theplasma to be in LTE[。]N ≥ 1.6 x 10 T (AE)。 (5)where△E (eV)is the largest observed transition ener-gY for which the LTE holds and T (K)is the excita-tion temperature，which equals to Te in LTE。For thelines used in this experiment，the largest AE is 5.5 eVfor Fe I 225.079 nm，and the electron temperature ob-served is approximately 8058 K. From Eq.(5)， theminimum electron density of 2.4×10 cm 。is requiredfor LTE to hold in the 1aser-ablated ron plasma. Asthe requirement limit is much lower than the electronnumber density that we obtained，LTE in the iron plasma2.5 Self-absorptionwavelength/rimFig.6 Stark broadened profile of Fe I 376.553 1"1112in laser-induced iron plasmais valid under present experimental conditions。

W hen the plasma parameters are evaluated by the Boltzmann plot method and Stark broadening，it is al-so important to verify that the plasma is not optically thick for the lines used 引.The spectral absorption canbe estimated from the following relation[ 。]忌 ( 。)-8.85×10- 。fJ j P， ( 。) (6)where k(cm )is the absorption coefficient，厂 is the absorption oscillator strength， 。(cm)is the wave-length，//i(cm-。)is the population density of the lower-level energy.P ( o)is the normalized line profile atthe center of the line and equals to 1/(7c△ 1/2).In the case of Fe I transition at 376.553 nm，the atomic den-sity is approximately equal to the electron density. Considering the values of the electron temperature and e-1ectron densitv (T 8058 K，N ～ 8.7 X 10 cm-。)，the estimated population density of the lower-level is2.5× 10 。cm～ ，and the corresponding absorption coefficient is 0.013 cm～ .W ith the measured plasmathickness D 1ess than 0.5 cm，the optical depth kD is much lower than 1.Therefore，the experiment is ear-ried out in optically thin plasma。

2.6 Other parametersIn 1aser ablation Dlasmas，the time for energy transfer from electrons to ions(10-如-10- s)is generallymuch shorter than the pu1se duration of typical ablation laser pulses(～ 10 ns) .A part of the laser beamwill be reflected or absorbed by the plasma formed j ust after the impact of the leading edge of the laser pulseon the surface. The plasma frequency is given as p-.8.9× 1 0。N . As the electron density increases，theplasma frequency will increase too. As soon as the plasma frequency becomes larger than the laser frequency，the1aser heam will be reflected. For a 1064 nm Nd：YAG laser，the corresponding frequency is 2.82× 10 Hzwhich is much 1arger than the plasma frequency 8.3× 10 Hz with the electron density obtained above·Thereore。the loss of energy due to reflection of the laser beam from the plasma can be assumed to be mslg-nificant[ 。

Although some species can be directly vaporized as ionized particles，plasma formation can be mainly as-crIbed to Drocesses involved in the laser-plasma interaction[ . There are two dominant photon absorptionDrocesses in the iron plasma. One is photoionization of excited or ground-state atoms· Considermg the Pho-ton energy of the 1064 nm 1aser(1.17 eV)is much lower than the excitation potential of iron (7·9 eV)㈦ ，the direct Dhotoionization by the absorption of a laser photon is ruled out.The only possibility for photoioni-zation absorption to occur is by simuhaneous absorption of at least seven photons，which is obviously less el-fieient for the 1064 nm laser。

- The second absorption mechanism is the inverse Bremstrahlung(IB)absorpti。nin w h tre el -trOns鼬 in kinetic energy from the laser beam and further promote iron plasma ionization and 。x t t 。 ·占-∞ -口 N- 暑 J0I1第 7期 Luo Wenfeng，et al：Measurements of iron plasma parameters produced by a 1064 nm pulsed 1695The IB absorption via free electrons is described as[21]- 1.37× 10 。N 。 (7)where (m)is the wavelength of the laser photons.W ith the experimental results，the IB absorption cod-ficient a1B is 0.14 cm- .Because of its 。dependence，the IB process is very efficient for the 1064 nm laser。

3 ConclusionIn this paper，laser ablation of iron sample using a 1064 nm pulsed Nd：YAG laser is performed in air atatmospheric pressure.A time delay of 5s is used to gate off the early part of the signal in order to avoid theintense initial continuum emission and improve the signal-to-background ratio.The electron number densityestimated from the Stark broadened profile of Fe I emission line at 376.553 nm is approximately 8.7× 10”cm 。，imP1ying that LTE under the present experimental condition is valid. The electron temperature isstudied using the improved iterative Boltzmann plot method and the value of 8058 K is inferred with final 33neutral Fe lines.W ith the experimental results，the plasma is verified to be under local thermodynamic equi-librium and optical thin conditions.The loss of energy due to reflection of the laser beam from the plasma isinsignificant because the plasma frequency(8.3× 10 Hz)is lower than the laser photon frequency (2.82×1 0 14 Hz).Experiments also show that during the laser-plasma interaction the dominant photon absorptionmechanism is the inverse Bremsstrahlung process，and its absorption coefficient is 0.14 cm。。

AcknowledgementsThe authors want to thank the researchers of the State Key Laboratory of Transient Optics and Photon-ics。Xian Institute of Optics and Precision Mechanics，Chinese Academy of Sciences for their experimentalsupport。

Re[1][2][3][4][5][6][7][82ferences：Kearton B，Mattley Y.Laser-induced breakdown spectroscopy：Sparking new applications[J].Nature Photonics，2008，2(9)：537-540Singh J P，Thakur S N.Laser-induced breakdown spectroscopy[M].Amsterdam：Elsevier Science BV，2006。

[9][1o][11][12][I3][14][15][16]ZhaoXiaoxia.Spatial distribution of electron temperature of laser-induced aluminum aloy E414d plasma[J].High Power Laser and ParticleBeams，2O11，23(9)：2519-2522。

Cremers D A，Radziemski L J.Handbook of laser induced breakdown spectroscopy[M].Chichester：John Wiley and Sons，Ltd，2006。

Aragon C，Aguilera J A.Characterization of laser induced plasmas by optical emission spectrOscopy：A review of experiments and methods[J].SpectrochimActa Part B，2008，63(9)：893-916。

Michel A P M.Review：applications of single shot laser-induced breakdown spectroscopy[J].SpectrochimActa Part B，2010，65(3)：185-191。

Pasquini C，Cortez J，Silva L M C et a1.Laser induced breakdown spectrosc。py[J].1 Braz Chem Soc，2007，18(3)：463-512。

Castle B C，Knight A K，Visser K，et a1.Batery powered laser induced plasma spectrometer for elemental determinationsLJ].J Anal AtomSpectrosc，1998，13(7)：589-596。

Milaen M，Laserna J J.Diagnostics of silicon plasmas produced by visible nanosecond laser ablation[J].Spectrochim Acta Part B，2001，56(3)：275-288。

Miziolek A W，Paleschi V，Schechter I.Laser-induced breakdown spectroscopy(IABS)：Fundamentals and applicationsrMJ.New York：Cambridge University Press，2006。

Amoruso S，Bruzzese R，Spinelli N，et a1.Characterization of laser-ablation plasmasJ].J Phys B：AtMol Opt Phys，1999，32(14)：R13l-Rl72。

National Institute of Standards and Technology.Atomic spectra database[DB/0L].http：//痙ata/asd.cfm。

Sabsabi M，Cielo P.Quantitative analysis of aluminum aloys by laser-induced breakdown spectroscopy and plasma characterization[J]。

Aydin U，Roth P，Gehlen C D，et a1.Spectral line selection for time resolved investigations of laser-induced plasmas by an iterative Boltz-mann plot method[J.Spectrochim Acta Part B，2008，63(10)：1060-1065。

Narayanan V，Thareja R K.Emission spectroscopy of laser ablated Si plasma reiated to nanoparticle formation[J.Applied s呖 fP Sci-ence，2004，222(1)：382-393。

Abdellatif G，Iman H.A study of the laser plasma parameters at diferent laser wavelengths[J].Sp ctrochimAf P B，2002，57(7)：1696 强 激 光 与 粒 子 束 第 25卷[17][18][19][20][21]1155-1165。

AguileraJ A，Aragon C.Characterization of laser-induced plasmas by emission spectroscopy with curvof-growth measurements.Part I：Temporal evolution of plasma parameters and sel abs0rption[J].SpectrochimActa Part B，2008，63(7)：784-792。

Le Drogof B，Margot J，Chaker M，et a1.Temporal characterization of femtosecond laser pulses induced plasma for spectrochemical analy-sis of aluminum aloys[J].Spectrochim Acta Part B，2001，56(6)：987-1002。

Amoruso S，Armenante M，Berardi V，et a1.Absorption and saturation mechanisms in aluminum laser ablated plasmas[J].Appl Phys A，1997，65(3)：265-271。

Harilal S S，Bindhu C V，Issac Riju C，et a1.Electron density and temperature measurements in a laser produced carbon plasma[J].ApplPhys，1997，82(5)：2140-2146。

Shaikh N M ，Hafeez S，Rashid B，et a1.Spectroscopic studies of laser induced aluminum plasma using fundamental，second and third har-monics of a Nd：YAG laser[J].EarPhys J D，2007，44(2)：371-379。

1064 am N d i YAG激光诱导铁等离子体特征参数的研究罗文峰 ， 赵小侠。， 朱海燕 ， 李冬冬 ， 李晓莉(1.西安邮电大学 电子工程学院，西安 710121； 2.西安文理学院 物理与机械电子工程学院，西安 710065)摘 要： 利用 1064 rim Nd：YAG激光器研究了激光诱导铁条等离子体的特征参数。为了减小测量误差和谱线自发辐射跃迁几率不确定性带来的计算误差，采用改进的迭代 Boltzmann方法精确求解铁等离子体的电子温度为 8058 K。Lorentz函数拟合 Fe I 376.553 nm得 到等离子体 的电子数密度为 8.7×10”cm～。分析表 明等离子体 的加热机制 主要是逆轫致 过程 ，其吸收系数是0.14 cm-1。实验数据证实激光诱导铁等离子体处于局部热力学平衙状态和光学薄状态。

关键词： 原子发射光谱； 激光诱导击穿光谱； 等离子体 ； 电子温度； 电子数密度