investigation of local thermodynamic equilibrium of laser induced al2o3 tic plasma in argon by spatially resolved optical emission spectroscopy

The correlation between the temporal behavior and the spatial temperature evolution are investigated under LTE condition for the possibility to use the temporal profile of Ti I emission

plasma in argon by spatially resolved optical emission spectroscopy

K Alnama, A Alkhawwam, and A K Jazmati

Citation: AIP Advances 6, 065112 (2016); doi: 10.1063/1.4954059

View online: http://dx.doi.org/10.1063/1.4954059

View Table of Contents: http://aip.scitation.org/toc/adv/6/6

Published by the American Institute of Physics

Investigation of local thermodynamic equilibrium of laser

optical emission spectroscopy

K Alnama, A Alkhawwam, and A K Jazmatia

Physics Department, Atomic Energy Commission, Damascus, Syria

(Received 25 February 2016; accepted 1 June 2016; published online 10 June 2016)

Plasma plume of Al2O3–TiC is generated by third harmonic Q-switched Nd:YAG nanosecond laser It is characterized using Optical Emission Spectroscopy (OES)

at different argon background gas pressures 10, 102, 103, 104 and 105 Pa Spatial evolution of excitation and ionic temperatures is deduced from spectral data analysis Temporal evolution of Ti I emission originated from different energy states is probed The correlation between the temporal behavior and the spatial temperature evolution are investigated under LTE condition for the possibility to use the temporal profile of

Ti I emission as an indicator for LTE validity in the plasma C 2016 Author(s) All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/) [http://dx.doi.org/10.1063/1.4954059]

I INTRODUCTION

Al2O3–TiC is one of the exceptional ceramic materials due to their high hardness and low wear rate and micromechanical properties that make it extremely advantageous for high precision parts in the manufacture of mechanical and electromechanical devices such as cutting tools mag-netic head sliders.1 3 Over the past twenty years, laser induced breakdown spectroscopy (LIBS) has emerged as valuable technique used in many applications in analytical, academic and govern-mental laboratories in different scientific and technological fields such as pulsed laser deposition (PLD) of thin films, laser treatment of surfaces, qualitative and quantitative elemental analysis, and nanotechnology.47

In ns-LIBS, a high-powered nanosecond laser pulse is focused on a solid, liquid, or gaseous sample, yielding a surface power density of order 108-1010 W/cm2 The free and loosely-bound electrons of the sample absorb energy from the laser pulse during hundreds of picoseconds through inverse Bremsstrahlung processes and release additional electrons through collisions The released electrons interact with atoms and molecules of the plume generating a high density of electrons The formed plasma lasts for microseconds during which atomic, ionic, and molecular emissions characteristic of the plasma can be measured.4,5

Different experimental conditions such as background gas pressure, sample material, wave-length and laser pulse energy could influence the spatial and temporal plasma formation and propagation Plasma parameters are dependent on these experimental conditions Excitation, ionic temperatures, electron density and effective lifetime are among these plasma parameters, and they are important to be known

Few works concerning study of the laser induced plasma of Al2O3-TiC target exist in the literature V Oliveira et al.8,9 explored the plasma plume of Al2O3-TiC created during KrF laser ablation They investigated by optical emission spectroscopy the spatially and temporally plasma distributions and correlated it with the surface morphology and composition of the target They found that the plume is enriched in Ti and depleted in Al by increasing the number of laser pulses

a Corresponding author Tel.: +963 11 2132580; fax: +963 11 6112289 E-mail address: pscientific5@aec.org.sy (A K Jazmati).

2158-3226/2016/6(6)/065112/11 6, 065112-1 © Author(s) 2016.

L D’Alessio et al.10analyzed the plasma obtained by laser ablation of titanium carbide targets

by time and space-resolved emission spectroscopy and intensified charge coupled device (ICCD) imaging They studied the effect of laser fluence on the different Ti ionization degrees and particles velocity They related the ablation mechanism with its influence on the film growth and found that

in the case of laser fluence higher than 3 J/cm2, the front of the emitting plume is identified with the presence of Ti2+ions.

M Mendes et al.11investigated mechanisms of formation of debris and particulates during mi-cromachining of Al2O3–TiC ceramics using pulsed 193 and 248 nm laser radiation They studied the influence of laser wavelength, fluence and ambient gas type, pressure and correlated it with changes

in the plume dynamics and particle formation behavior They found that the amount of debris accumulated increases with increasing gas atomic/molecular weight and pressure They correlated that with the correspondent increase in the plume confinement by the atmosphere, which enhances collisions in the plume, promoting the further growth of particles

The temporal profile of the plasma emission has an essential effect of the analytical use of LIBS

It depends on the population of the upper excited states of the radiative atoms in the plasma whether they are subject to LTE condition or not The easiest way to estimate the LTE condition validity is McWhirter criterion Different authors have suggested that this LTE criterion is necessary but not

sufficient condition and the mere use of it in order to prove the existence of LTE should be aban-doned.12,13Diwakar and Hahn14have concluded that the plasma reaches LTE conditions only after

an initial delay of about 20 – 40 ns following plasma inception Harilal et al.15have mentioned that at larger distances and later times, the densities of the plasma are low and the movement of the boundary region is rapid, so LTE is probably not a good assumption Colao et al.16have studied the role of the laser energy where the LTE condition holds, at atmospheric pressure, and they have concluded that

as the laser energy increases, the temporal window, where the LTE condition is justified, increases Thiyagarajana and Scharer17have shown that the plasma of air is in LTE condition for time less than

2 microseconds The same conclusion has been found by Herrera et al.18by studying Al alloy laser induced plasma in air generated by Nd:YAG laser Barthelemy et al.19have studied the Al plasma by excimer laser in air, they found some evidence of a departure from LTE for the ionized states, presum-ably because the plasma cannot be considered stationary for the higher energy states Capitelli et al.20

have mentioned that LIBS plasmas fulfill LTE condition, during the measurement time, even though the plasma parameters (temperature, electron density Ne) rapidly change due to plasma expansion Therefore, it is important to determine the conditions for expanding high density plasmas to be in an equilibrium state as well as of the time duration for the existence of such equilibrium

In this work, we investigate Al2O3–TiC plasma characterization, via different parameters, such

as excitation and ionic temperatures and electron density as well as their spatial evolution In addition, we explore the temporal profile of Al2O3–TiC plasma emissions of Ti I selected lines at

different argon background pressures These profiles could demonstrate the role of excited state in the plasma dynamics The combination between the excitation and ionic temperatures with the tem-poral profiles may present a helpful tool to elucidate the LTE validation and the plasma dynamics This study could be helpful to improve the laser micromachining efficiency for ceramic materials

II EXPERIMENTAL SETUP

The experimental setup is shown in Fig 1 A Q-switched Nd:YAG laser (Quantel YG series 820) delivers 355 nm pulses at 10 Hz with pulse duration of 10 ns (FWHM) The laser beam energy

is fixed at 30 mJ The 9 mm diameter laser beam was focused through a high power plano-convex lens with focal length f= 75 mm and a quartz window inside the PLD chamber onto the rotated

Al2O3–TiC target at 45◦ from the normal target surface The targets used in this work consist of

66 wt % of alumina and 34 wt % of titanium carbide

The PLD vacuum chamber was equipped with a system for gas feeding and pumping out with

a pressure controlling system and rotated target holder in order to provide a fresh surface to avoid errors associated with local heating and drilling of the target Laser energy was measured using a pyroelectric smart sensor energy meter (LM-10 LP Coherent Labmaster), a good reproducibility was achieved and laser output energy fluctuations were below 5 %

FIG 1 Schematic diagram of the experimental setup.

The laser fluence at the focused spot area on the target surface was estimated to be 15 J cm−2 which leads to a pulse power density of 1.5 GW cm−2 A pulse delay generator (Stanford DG 535) was used to synchronize both laser and CCD camera

The plasma light emitted from the plume was optically imaged 1:1 onto the optical fiber bundle entrance at right angle to the normal of the target surface (see Fig.1) Imaging was done by using a confocal optical system consists of a diaphragm and two quartz lenses (f1= f2= 10 cm) The emis-sion light of the plasma was dispersed by Czerny-Turner monochromator (HORIBA JOBINYVON

1250 M series II) with 1800 g.mm−1grating blazed over the spectral range 450-850 nm and directed onto the CCD camera (LINESCAN, TOSHIBA 3648×1), or the PMT (Hamamatsu E678-11A) entrances to detect the dispersed emission The entrance slit width of the monochromator was set at

50 µm

The temporal profiles have been recorded by PMT coupled with Tektronix digital oscilloscope (TDS 3054C) Each temporal profile is an average of 128 ones to improve signal to noise ratio The spatial study was done by moving the entrance of the optical fiber bundle (mounted on

X, Y, Z micro-displacement) in the plane of the image in order to move along the plasma axial direction

III RESULTS

The emission spectrum of Al2O3–TiC plasma was recorded in the range of 450-820 nm It is dominated by the emission of Ti I neutral atoms and Ti II ions In addition, three O I and two Ar I lines have been detected in the spectrum There is no clear emission from carbon or aluminum in the spectrum because all the persistent lines of theses neutral atoms are outside our spectral range Several detected lines cover a wide range of Ti I excited states, Ti II, O I and Ar I are listed in tableI

TABLE I Di fferent detected Ti I, Ti II, O I and Ar I lines, energy state, configuration, term and j for upper and lower levels.21

Wavelength

(nm)

Upper Level (cm−1)

Lower Level (cm−1)

Upper Level configuration term j

Lower Level configuration term j

Ti I 469.13 29 912.286 8 602.3441 3d2( 3 F )4s4p( 1 P◦) w3D◦ 3 3d24s2 a3P 2

Ti I 477.82 38 959.507 18 037.213 3d 3 ( 2 G )4p x 1 G ◦ 4 3d 3 ( 2 H )4s a 3 H 4

Ti I 479.98 39 115.957 18 287.554 3d3( 2 H)4p x3H◦ 4 3d3( 2 G)4s b1G 4

Ti I 487.01 38 668.887 18 141.265 3d 3 ( 2 H )4p z 3 I ◦ 6 3d 3 ( 2 H )4s a 3 H 5

Ti I 489.99 35 559.648 15 156.802 3d3( 2 G )4p y3H◦ 5 3d3( 2 G )4s a3G 4

Ti I 492.18 37 852.437 17 540.188 3d 3 ( 2 D2 )4p u 3 F ◦ 4 3d 3 ( 2 D2 )4s a 3 D 3

Ti I 511.10 45 097.323 25 537.286 3d24s ( 2 D )5s 3 D 2 3d2( 1 D )4s4p( 3 P ◦

) 3 P◦ 1

Ti I 512.04 40 319.829 20 795.603 3d 3 ( 2 H )4p z 1 I ◦ 6 3d 3 ( 2 H )4s a 1 H 5

Ti I 515.22 19 573.974 170.1328 3d2( 3 F )4s4p( 3 P ◦

) z3F◦ 4 3d24s 2 a3F 3

Ti I 517.37 19 322.984 0.000 3d 2 ( 3 F )4s4p( 3 P ◦

) z 3 F ◦ 2 3d 2 4s 2 a 3 F 2

Ti I 519.30 19 421.580 170.1328 3d2( 3 F )4s4p( 3 P ◦

) z3F◦ 3 3d24s2 a3F 3

Ti I 529.58 27 480.066 8 602.3441 3d 2 ( 1 D )4s4p( 3 P ◦

) x 3 D ◦ 3 3d 2 4s 2 a 3 P 2

Ti I 551.25 29 912.286 11 776.812 3d2( 3 F )4s4p( 1 P ◦

) w3D◦ 3 3d3( 4 F )4s b3F 4

Ti I 556.55 36 000.148 18 037.213 3d 2 ( 1 G )4s4p( 1 P ◦

) y 1 G ◦ 4 3d 3 ( 2 H )4s a 3 H 4

Ti I 564.41 36 000.148 18 287.554 3d2( 1 G )4s4p( 1 P ◦

) y1G◦ 4 3d3( 2 G )4s b1G 4

Ti I 582.37 35 454.052 18 287.554 3d 3 ( 2 G )4p y 3 H ◦ 4 3d 3 ( 2 G )4s b 1 G 4

Ti II 476.28 29 734.6206 8 744.3406 3d2( 3 F )4p z4G◦ 7/ 2 3d2( 1 D )4s a2D 5/ 2

Ti II 491.12 45 548.9273 25 192.9650 3d 2 ( 3 P )4p y 2 P ◦ 3 / 2 3d4s 2 c 2 D 5 / 2

Ti II 538.10 31 207.5111 12 628.8455 3d2( 3 F )4p z2F◦ 5/ 2 3d3 b2D2 3/ 2

O I 777.19 86 631.454 73 768.200 2s 2 2p 3 ( 4 S ◦

)3p 5 P 3 2s 2 2p 3 ( 4 S ◦

)3s 5 S ◦ 2

O I 777.42 86 627.778 73 768.200 2s22p3( 4 S ◦

)3p 5 P 2 2s22p3( 4 S ◦

)3s 5 S◦ 2

O I 777.54 86 625.757 73 768.200 2s 2 2p 3 ( 4 S ◦

)3p 5 P 1 2s 2 2p 3 ( 4 S ◦

)3s 5 S ◦ 2

Ar I 763.51 106 237.5518 93 143.7600 3s23p5( 2 P ◦

3/2 )4p 2 [ 3 / 2 ] 2 3s23p5( 2 P ◦

3/2 )4s 2 [ 3 / 2 ] ◦

2

Ar I 811.53 105 462.7596 93 143.7600 3s 2 3p 5 ( 2 P ◦

3/2 )4p 2 [ 5 / 2 ] 3 3s 2 3p 5 ( 2 P ◦

3/2 )4s 2 [ 3 / 2 ] ◦ 2

A Spatial behavior of excitation and ionic temperatures

The excitation and ionic temperatures have been calculated at different positions along the plume axe for pressures in the range (10 Pa – 105Pa) of argon background gas They are calculated

by using Boltzmann plot method employing the emission lines at 469.13, 487.01, 489.99, 492.18, 515.22, 517.37, 519.30, 529.58, and 551.25 nm for Ti I and at 476.28, 491.12 and 538.10 nm for Ti

II These lines present wide range of upper state energies and they are well isolated The spectro-scopic constants of these spectral lines have been taken from Ref.21 Fig.2represents the spatial behavior of the excitation and ionic temperatures in the range (10 Pa – 105Pa) of argon pressure If the LTE is verified in the plasma, the excitation and ionic temperatures are almost equal We notice from Fig.2that at 10 Pa argon pressure, the discrepancy between the two temperatures is spatially increasing, while at 102Pa, the difference between the two temperatures is quasi-stable In these two pressures, the ionic temperature (Tion) is higher than the excitation temperature (Texc) over all the studied spatial range At 103Pa, Tionis still higher than Texcup to 8 mm and then Tionbecomes lower than Texc At higher Ar pressures (104, 105Pa), the two temperatures are almost equal

The difference between the ionic and excitation temperatures has been explored temporally in the delay time range (500 to 5000 ns) by Galmed et al.12for different laser energies at atmospheric pressure They concluded that for low power density (2 × 1011 W/cm2), the ionic temperature is higher than the excitation temperature while at high power density (8 × 1011 W/cm2) the two temperatures are quasi-equal and according to their results LTE condition is verified over all the temporal range

Generally, the excitation and ionic temperatures have been found to differ more in the initial stage of the expansion than at later times Radziemski et al.22have compared the two temperatures temporally at atmospheric pressure and have mentioned that the two temperatures are quasi equal and the plasma is in LTE after the first 1µs Although that the laser power density used in this

FIG 2 Spatial evolution of the excitation (filled square) and ionic (open square) temperatures at argon gas pressures: 10 (a),

10 2 (b), 10 3 (c), 10 4 (d), 10 5 Pa (e).

work is much lower than that used by Galmed et al.,12 our results about the two temperatures at atmospheric pressure are indicating that the LTE is verified at all distances which is similar to the result found by Galmed

Grojo et al.23 have investigated the temporal evolution of the femtosecond laser induced Ti plasma temperature on a nanosecond time scale They have mentioned that the temperature de-creases from T ≈ 1 × 104 to 5 × 103K within the first nanoseconds This behavior was attributed

to a quasiadiabatic expansion process during which the thermal energy is converted into kinetic energy They have also showed that the temperature decreases more slowly for the late expansion stage, during which the dominant energy-loss mechanisms are ascribed to the radiative losses and heat exchange with the low-pressure atmosphere

The spatial behavior of Texcat different argon pressure is similar to our result for a Yb plasma.24

In addition, the ionic temperature approaches the excitation temperature by increasing the argon background pressure At 10 Pa, Tionis spatially increasing and diverted from Texc, which could be attributed to the violation of LTE condition in the plasma However, the close temperatures values

at 2 mm indicate the existence of LTE near the target surface In addition, this increase of Tionwith increasing distance, as shown in Fig.2, could be explained by a spatial increase of the ionization rate coefficient,25 such a spatial increase of Tion disappears at high pressure By increasing the argon pressure, Tionapproaches Texcwhich means that LTE becomes more valid Moreover, as it is mentioned by Camacho et al.26the real plasmas are an approximation to LTE and there is always

a spatial inhomogeneity in the means of temperature so LTE is applied only within a small volume

of the plasma To our knowledge, only two works concerning the spatial behavior of the ionic temperature are accomplished.27,28The first one has been done on silicon sample at atmospheric pressure where the studied spatial range was up to 0.3 mm, the behavior was continuously decreas-ing However, no comparison has been done with the excitation temperature The second work has been done by De Giacomo et al.28where KrF excimer laser is utilized to ablateTiO2target at 1 Pa background pressure of O2 The ionic temperature was higher than the excitation temperature over all the spatial range, this difference was explained as an effect of radiative processes

B Spatial behavior of electron density

Electron density (Ne) has been estimated from the full width at half maximum (FWHM) of the Stark broadened profile of neutral oxygen line at 777.19 nm using the following relation:

ω= 2ωp

(

Ne

1016

)

(1) Where ω is the FWHM of the considered oxygen transition and ωp is the Stark broadening parameter (ωp= 0.00228 nm).29The FWHM of raw spectral data was corrected for instrumental broadening (0.02 nm), while pressure and Doppler broadenings were ignored

The spatial distribution of electron density at different pressures (10, 102, 103104and 105Pa) shows a typical decreasing evolution as shown in Fig 3 At 10 Pa and a distance of 2 mm, the electron density is high and the collisional process becomes more dominant than the radiative mech-anism, hence, the LTE condition could be valid to describe the plasma.12 , 30Farther than 2 mm the electron density decreases exponentially and does not follow 1/z law which indicate that the plasma does not propagate freely and the expansion of the plasma plume is not one dimensional as it is considered in other works.31 , 32Therefore, the radiative process starts to compete with the collisional one and the collisional-radiative model (CR) could be better to describe the plasma At 102Pa Ne

increases slowly up to 13 mm While for 103 Pa, it almost increases until 5 mm then it starts to decrease Similar behavior was already seen by Coons33at 103Pa of air background pressure where

Neincreases up to 1 mm from the target surface, and then it decreases At high pressure 104, 105Pa, the plasma is confined and its spatial propagation is limited, therefore the spatial distribution of Ne

is continuously decreasing in a similar behavior of other works.34 – 38

At distances up to 5 mm, one can notice from Fig.3that Nedecreases with Ar pressure except for 102Pa where the value of Neis slightly increased at larger distances Coons has shown that at

103Pa argon pressure there is a sudden increase of Nein comparison with other pressures.33This spatial behavior of Neat 102Pa in our work needs more extensive research with smaller pressure variation and it will be examined in future work In addition, similarity between the spatial behavior

of Neand line intensity is noticed in this work (not shown here) Therefore, we can assume that a good number of excited atoms in the plasma originated from recombination processes According to Capitelli et al.20a quasi-equilibrium state is achieved for laser induced plasma, during the measure-ments which are made in the microsecond time scale, i.e the plasma is under typical recombination conditions In this state, typical electron densities are in the interval 1015≤ Ne ≤ 1018cm−3, these values are strongly depending on the delay time from the plasma formation

According to McWhirter criterion39which constitutes a necessary but not a sufficient condition for LTE, a minimal electron density value of Ne= 2 × 1015cm−3is found for 19421 cm−1excited state assuming a typical plasma temperature of T= 7000 K One can notice from Fig.3 that the measured electron densities at all distances are larger than this value

FIG 3 Spatial behavior of electron density at di fferent argon pressure.

G Travaillé et al.40 have developed a CR model to describe the plasma, and showed that the McWhirter criterion well agrees with the threshold given by the CR model They have also mentioned that, at low energy states, LTE is considered as a good model for describing the plasma They showed that the electron density threshold calculated for the ions is more restrictive than the neutral atoms by one order of magnitude Therefore, it is more difficult to achieve LTE criterion for the ionic states in the plasma In addition, according to Grojo et al.23 in LIBS experiments at atmospheric pressure the laser induced plasma is strongly confined and consequently the temporal behavior of the plasma density decreases slowly and the LTE criterion is fulfilled during a larger time interval On the other hand, this behavior decreases much faster and the LTE criterion is only satisfied for a shorter time under low-pressure background gas

We present in the next section the possibility to use the temporal profile as an indicator for LTE existence in the plasma

C Time dependence Profiles

As the pulse laser energy is transferred to the target surface within 10 ns, which is the width of the laser pulse, plasma plume is formed and many of the energy states are populated Fig.4shows

an example of the temporal profile of Ti I emissions, it consists of three regions In first region the continuum emission, due to the inverse Bremsstrahlung effect, dominants for about 300 ns After that and due to collisions between plasma species, the excited state levels start to be repopulated (second region) which influences the decay time of the emission (third region)

Energy states, pressures of the surrounding gas and observation distances of the plasma plume are important parameters that can influence the temporal profile.24 In order to study the effect of these parameters, a comparison between the recorded temporal profiles is helpful, as it will be discussed below, where the signal of the continuum emission (300 ns) is omitted

1 Energy state effect

In order to study the energy state effect on the temporal profiles, two diverse energy states

45093 cm−1(high) and 19421 cm−1(low) are chosen Fig.5shows both profiles recorded at 10 Pa argon gas pressure and an observation distance of 3 mm The two profiles show similar temporal evolution with 200 ns time delay between their maximum intensities This delay could originate from the depopulation of higher states to lower states.41It is clearly noticed that the temporal profile

of the lower state starts to decrease when the higher state is already depopulated

FIG 4 Temporal profile (O-TOF) of Ti I emission from the 19421 cm−1energy state (519.3 nm) at 10 3 Pa and observation distance of 3 mm.

FIG 5 Temporal profiles for two energy states 45097 cm−1and 19421 cm−1at 10 Pa and observation distance of 3 mm.

The temporal profiles for three different energy states (45097 cm−1, 36000 cm−1and 19421 cm−1 corresponding to 511.10, 564.41 and 517.37 nm respectively) show different shapes at 103Pa argon pressure and at 5 mm as it is shown in Fig 6, where a second peak appears with a time delay of approximately 4 microsecond after the mean peak This peak could be originated from plume–gas interaction that increases more clearly by increasing the background pressure Harilal et al.41have showed that, the temporal profile of excited Sn+ changes into double peaks distribution as the

background argon pressure increases They have also mentioned that the double peak distribution existed only in a certain pressure range, indicating that argon background gas plays an important role in the formation of the first peak Therefore, it is important to explore the pressure effect on the temporal profiles of the observed transitions, as it will be discussed below

It is noticeable from Figs.5and6that there is a broadening effect on the temporal profiles by decreasing the energy states and increasing the argon pressure from 10 Pa to 103Pa This behavior might be correlated with the one of the excitation and ionic temperatures (Fig.2), where the di ffer-ence between ionic and excitation temperatures decreases as the argon pressure increases indicating

a transfer from non-LTE to LTE condition This broadening is less effective for high excited states, which could be an indicator that it is difficult to conserve the equilibrium in the plasma for the high energy excited states, and the LTE condition is valuable only for the low and mid-energy excited states which agrees with Barthelemy et al work.19

In order to calculate the effective life time of the excited state in the plasma, the third region

of temporal profile has been fitted using an exponential equation.42 , 43An example of this fitting is shown in Fig4 This fit is applied for all observed temporal profiles and subsequently their effective lifetimes are deduced for 9 different detected lines (511.21, 512.04, 479.98, 550.39, 564.41, 491.36, 589.93, 519.30, 517.37) from tableIcover a wide range of Ti I upper excited states Fig.7shows

FIG 6 Comparison between temporal profiles for three di fferent energy states at 10 3 Pa and observation distance of 5 mm.

FIG 7 Ti I e ffective lifetime via the upper excited state energy at 10 Pa and observation distance of 3 mm.

the dependence of the effective lifetime on the upper excited state energy corresponding to temporal profiles recorded at 10 Pa and observation distance of 3 mm

Under the LTE condition in the plasma, the collisional effect between plasma species is domi-nant and therefore all states, theoretically, must have the same depopulation decay time as long as the plasma temperature changes adiabatically Since there is a clear difference between the effective lifetimes deduced from the depopulation decay for different excited states as shown in Fig.6, i.e LTE might be absent in our Ti studied plasma at 103 Pa Hence, the temporal behavior of the depopulation of excited state atoms follows a collisional-radiative model rather than LTE condition

as it is given by equation (2)

dn

dt =

Rrad+

Where Rradand Rcolldenote all possible radiative and collisional excitation-deexcitation events.30,40

It is clear that the low energy excited states have a high effective lifetime indicating that they were partially populated by the depopulation of higher energy excited states as it is mentioned above (see Fig.5) Otherwise, at 104Pa of argon pressure, the decreasing behavior of the effective lifetime via upper energy states (not presented) is not any more satisfied indicating that the LTE condition is more probably fulfilled

In addition, another explanation of effective life time decreasing with energy states (fig.7) could be related to the temperature decreasing with time (in condition of the plasma still maintains LTE) Therefore, the decay time for the higher levels will be shorter than the lower energy levels

We can conclude that temporal profiles of different transitions originating from different excited states might be considered as an indicator of LTE existence in laser induced plasma

2 Pressure effect

In order to study the pressure effect on the temporal profile, an observation distance and energy state have been fixed at 3 mm and 36000 cm−1respectively Fig.8shows a comparison of the tem-poral profiles at different pressures (10, 102, 103and 104Pa) The temporal evolution of the profiles for the low pressure (10, 102Pa) shows similar behavior and their depopulation decay decreases quickly within almost 1 microsecond While by increasing the argon gas pressures, the mean peak

of the temporal profiles gets broaden from 350 ns at 10 Pa to reach 2300 ns at 104Pa

In addition, a temporal stretching of the depopulation decay is observed and lasts for more than

10 microseconds, resulting in an increase of the effective life time of the emissions This effect could be explained by increasing the collisions between plasma species with the surrounding argon gas at high pressures.41