journa l homepage: www.e lsev ier .com/ locate /apsusc
omparison of the laser ablation process on Zn and Ti using pulsed digitalolographic interferometry
. Amera,∗, P. Grena, A.F.H. Kaplana, M. Sjödahla, M. El Shaerb
Department of Applied Physics and Mechanical Engineering, Luleå University of Technology, SE-971 87 Luleå, SwedenDepartment of Engineering Physics and Mathematics, Faculty of Engineering, Zagazig University, Zagazig, Egypt
r t i c l e i n f o
rticle history:eceived 7 December 2009eceived in revised form 11 February 2010ccepted 20 February 2010vailable online 26 February 2010
a b s t r a c t
Pulsed digital holographic interferometry has been used to compare the laser ablation process of a Q-switched Nd-YAG laser pulse (wavelength 1064 nm, pulse duration 12 ns) on two different metals (Zn andTi) under atmospheric air pressure. Digital holograms were recorded for different time delays using colli-mated laser light (532 nm) passed through the volume along the target. Numerical data of the integrated
eywords:aser ablationetals
ulsed digital holographic interferometry
refractive index field were calculated and presented as phase maps. Intensity maps were calculated fromthe recorded digital holograms and are used to calculate the attenuation of the probing laser beam bythe ablated plume. The different structures of the plume, namely streaks normal to the surface for Zn incontrast to absorbing regions for Ti, indicates that different mechanisms of laser ablation could happenfor different metals for the same laser settings and surrounding gas. At a laser fluence of 5 J/cm2, phase
he abism.
explosion appears to be tbe the dominant mechan
. Introduction
Laser ablation is the process of removing material from a sur-ace as a result of irradiation with a laser beam. It has a lot ofpplications especially in industry including microstructure modi-cation of materials [1,2], chemical analysis [3], microhole drilling4], art conservation [5] and thin film deposition [6]. The physicalrocess of laser ablation and laser-induced plume formation is notully understood. The process depends on the optical and thermalroperties of the target [7], the laser parameters [8–10] and thembient gas conditions [11,12]. There are several mechanisms ofaser thermal ablation, namely normal vaporization, normal boil-ng and explosive boiling (phase explosion). The thermal ablationrocess with laser pulse duration in the ns range can be described
n three different stages [13]. At the first stage, the laser light strikeshe solid and is absorbed by the electrons in the solid. After a periodf tens of ps the excited electrons undergo electron–phonon relax-tion and the energy is transferred to the lattice. Through latticeibrations, the transferred energy is dissipated from the irradiatedone to the bulk in the form of heat which results in melting of
he surface layer. At this stage, laser–solid and laser–liquid inter-ctions are dominant. At the second stage, the material from theeated volume is ejected but continues to absorb energy from the
aser, resulting in the formation of plasma in front of the surface.
At this stage, laser–gas or laser–plasma interactions are dominant.The third stage begins after the termination of the laser pulse. Herethe plume expands adiabatically in three dimensions. If the abla-tion process takes place in a background gas, the plume compressesthe surrounding gas and forms a shock wave. This mechanismis called normal vaporization. Normal boiling requires that thepulse duration is sufficiently long for heterogeneous vapor bubblenucleation to occur. If the laser fluence (energy/cm2) is sufficientlyhigh and the pulse length is sufficiently short the surface canreach a temperature higher than the normal boiling point, result-ing in a superheated, metastable state. As the surface temperaturereaches 0.9Tc, where Tc is the thermodynamic critical temperature,homogenous bubble nucleation occurs and the target makes a rapidtransition from superheated liquid to a mixture of gas and liquiddroplets leaving the target like an explosion. This mechanism isknown as phase explosion [14,15]. Several studies have demon-strated evidence of phase explosion induced by a nanosecond laserpulse on metallic targets [16–18].
The laser ablation induced plume and the formation of the shockwave have been studied by several authors including the use ofshadowgraphy [12], interferometry [19], schlieren [20] and a probebeam deflection technique [21]. The previous techniques give infor-mation about the transmission of the probe beam or about the
refractive index change along its path. Using pulsed digital holo-graphic interferometry both information about the amplitude andthe phase of the probing beam can be stored in the digital hologramsfrom which the attenuation of the probe beam by the inducedplume as well as the refractive index change along its path can
e calculated [22]. In this paper laser ablation of two differentetals (Zn and Ti) with significant differences in their physical
roperties has been studied using pulsed digital holographic inter-erometry. The comparison of the phase maps and the intensity
aps of the two metals shows that different ablation mechanismsan be observed for the same laser settings and surrounding gas.ection 2 of this paper describes the experimental set-up used forhe experiments and the way information is extracted from the
easurements. Section 3 presents the main results obtained forhe two metal surfaces at different time instances and in Section 4he characteristics of the results are compared and discussed. Theaper ends with a few concluding remarks.
. Experimental method
A top view of the experimental set-up is shown in Fig. 1(a). Annjection-seeded, twin oscillator, Q-switched Nd:YAG laser systemSpectron SL804T) is used as light source. Each laser comprises a sin-le oscillator with a single power amplifier in series. Each oscillators configured with a telescopic resonator with an intracavity mode-ontrolling aperture. This gives rise to a true TEMoo spatial profileor spatial uniformity and coherence. Since the two laser oscillatorsre seeded from the same stabilised CW Nd:YAG laser the pulsesrom the two oscillators are coherent. The laser system operates at0 Hz but the time separation between the pulse trains from thewo lasers can be set from zero to any time. For reliable seeding, it
s necessary that the oscillators are run repetitively. Stable singlehot operation is not possible. Instead, fast solenoid-activated beamump shutters allow access to a single, stable, single-frequencyulse. The fundamental Nd:YAG wavelength 1064 nm from oneavity is used for processing and the frequency doubled 532 nm
ig. 1. The experimental set-up. (a) Top view of the set-up. M1, M2: mirrors, NL:egative lens, L1: focusing lens, L2: collimation lens, L: lens system for imaging, A:perture, D: diffuser, BS1 and BS2: beam splitters, R: reference beam, O: object beam.b) 3D sketch of the target and the coordinate system.
nce 256 (2010) 4633–4641
pulse from the second cavity is used for the measurement. Thegreen light from the Nd:YAG laser is split by a beam splitter (BS1).The reflected part is reflected by mirror M1, expanded by a neg-ative lens (NL), collimated by a positive lens (L2) and illuminatesa diffuser (D) after passing along the target. The light that passesthe beam splitter BS1 is used as reference beam (R) and it is guidedthrough a fibre optic cable to the beam splitter BS2 from where itilluminates the CCD-detector. The camera is a PCO Sensicam dou-ble shutter, with a resolution of 1280 × 1024 pixels, a pixel size of6.7 �m × 6.7 �m and a dynamic range of 12 bits. The camera is com-puter controlled via a fibre optic cable and externally triggered tobe synchronised with the laser pulses. The diffuser is imaged ontothe CCD-detector by a two-element lens system (L); each element isa plano-convex lens with a focal length of 100 mm. An aperture (A)with a size of 2.45 mm × 2.45 mm is placed between the two ele-ments of the lens system. The field of view is 3.65 mm × 2.92 mm.The tip of the optical fibre is positioned in such a way that seen fromthe detector it should appear to come from the same plane as theaperture and one aperture width (2.45 mm) from the edge of theaperture. In this way the interference pattern between the objectand reference light is spatially separated from the object light selfinterference term in the Fourier domain, which enables the Fouriertransform method to be used for the calculation of the complexamplitude [23]. Fig. 1(b) shows a 3D sketch of the target and thecoordinate system. The X and Y axes are in the plane of the targetand the Z axis is pointing outward.
An IR pulse (� = 1064 nm and pulse duration = 12 ns) from theNd:YAG laser is focused to a 35 �m diameter beam waist by a80 mm focal length lens. The spot diameter has been increasedto 0.68 mm by adjusting the distance between the focusing lensand the target to 62 mm. The distance from the processing areato the diffuser is chosen as short as practically possible (6 mm) tominimize the effect of bending of the probing light.
The experimental procedure is as follows: Two digital holo-grams of the diffuser are recorded. The first image (reference image)is recorded with the processing beam blocked, thus recording theundisturbed air. The second image (deformed image) is recordedwith the processing beam on and it contains information about thedisturbed volume at a certain time between the two laser pulses.The complex amplitudes are then calculated from the reference anddeformed image, respectively. Since the setup is stable and the lasersystem has a high degree of coherence, it is possible to compare dif-ferent recordings. In particular the interference term, W, betweenthe deformed and the reference images may be calculated as,
W = U2U1∗, (1)
where U2 is the complex amplitude of the deformed image, U1 is thecomplex amplitude of the reference image and * denotes complexconjugate. The field W given by Eq. (1) is in general a complex fieldwhose magnitude represents the intensity in the image and whosephase gives the phase change between the two recordings. To allowquantitative comparisons the intensity image we use is defined as:
I =∣∣W∣∣∣∣U1
∣∣2, (2)
where the normalization is introduced to reduce the effect of speck-les. The intensity may then vary between zero and two due topossible absorption and interference. The phase difference ��between the deformed and the reference images is calculated as:
(Im(W)
)
�� = arctan
Re(W), (3)
which results in a wrapped phase map. An unwrapping proce-dure is finally applied to remove possible 2� ambiguities. Moredetails about the procedure to obtain the phase data are presented
ce Scie
itaTdis3aT
3
s
3
fliteapls
Fa
E. Amer et al. / Applied Surfa
n [24]. With reference to Fig. 1, W and hence the intensity andhe phase difference measured, will become the integrated effectlong the propagation path, l, through the measurement volume.he phase difference caused by propagating the beam a distancel is k · �n · dl, where k is the wave number and �n is the changen refractive index. If the disturbance is assumed to be rotationallyymmetric, the Radon inversion method can be used to obtain theD field from the integrated field (the 2D map). In the following theblation plume images from two different materials, namely Zn andi (under equal pulsing conditions), are evaluated and compared.
. Results
In this section the results from Zn and Ti experiments will behown.
.1. Zn results
Time series of intensity maps and phase maps of Zn at a laseruence of 5 J/cm2 (beam diameter: 0.68 mm, pulse duration: 12 ns)
s shown in Figs. 2 and 3, respectively. Fig. 2 shows that the struc-ure of the induced plume is different for different time stages. At
arlier time (t = 50 ns) the intensity map shows that a dark regionppears close to the target, see Fig. 2(a), which indicates that therobing beam is nearly fully absorbed by the induced plume. At
ater times from 200 ns to 800 ns, Fig. 2(b)–(e) shows that there aretreaks appearing as dark and bright lines close to each other, which
ig. 2. A time series of intensity maps of Zn at a laser fluence of 5 J/cm2. (a) t = 50 ns, (b) t = 2nd bright represents high transmission.
nce 256 (2010) 4633–4641 4635
indicates deflection of the probing beam in nearly straight linesout from the target. The distance between the outermost streaksis roughly 0.8 mm which is in the same order of magnitude as theprocessing beam spot diameter. The extinction of the probe beam(532 nm) taken as an average from Z = 0.085 mm to Z = 0.2 mm att = 50 ns and t = 600 ns as a function of Y is shown in Fig. 4. We havechosen to define the extinction, E, as E = 1 − I, where I is the inten-sity defined by Eq. (2). Fig. 4 shows that the probe beam is almostcompletely absorbed (about 70%) at early time (t = 50 ns), while atlater time (t = 600 ns), positive and negative values for the extinc-tion close to each other (where the streaks appear in the intensitymap) can be seen (therefore we avoid the term absorption, as speck-les, diffraction and noise also contribute). That indicates deviationand hence diffraction of the probe beam at the positions of thestreaks. On average from Y = 0.5 mm to Y = 2.5 mm the extinction ispositive and it is about 10%.
The streaks are even more clearly seen in the correspondingphase maps, see Fig. 3(b)–(e). In Fig. 3, the phase differences varybetween −7.4 rad (black) and 7.4 rad (white) where the grey regionoutside the shock wave front represents zero phase difference. Fig. 3shows that the phase maps of Zn are asymmetric; hence the Radoninversion cannot be used to calculate the 3D refractive index field.
Instead the integrated refractive index will be used to get a qual-itative indication about the density of the ablated plume. Phasedifference profiles at different distances from the Zn surface ata laser fluence of 5 J/cm2 and a time delay of 600 ns taken fromFig. 3(d) are shown in Fig. 5. The figure shows that the phase dif-
00 ns, (c) t = 400 ns, (d) t = 600 ns and (e) t = 800 ns. Dark represents low transmission
4636 E. Amer et al. / Applied Surface Science 256 (2010) 4633–4641
2. (a)
fitst
Ffl
Fig. 3. A time series of phase maps of Zn at a laser fluence of 5 J/cm
erence at the positions where the streaks appear in the phase map
s greater than zero and decreases with increasing distance fromhe target. This is an indication that the refractive index of thetreaks is larger than the undisturbed refractive index and hencehe density is larger than the density of the undisturbed air (from
ig. 4. The extinction of the probe laser beam (532 nm) by the Zn plume at a laseruence of 5 J/cm2 at t = 50 ns and t = 600 ns, average from Z = 0.085 mm to Z = 0.2 mm.
t = 50 ns, (b) t = 200 ns, (c) t = 400 ns, (d) t = 600 ns and (e) t = 800 ns.
the Gladstone–Dale equation [25]). It may then be concluded thatthe streaks consists of optically denser material (metal matter in
some state) as compared to air. A disturbance can be noticed inthe upper part in both intensity maps, Fig. 2(b)–(d) and in the cor-responding phase maps, Fig. 3(b)–(d). It is labelled by the letter Iin Fig. 2(c). The cause of this disturbance was however not inves-
Fig. 5. Phase difference profiles of Zn at a laser fluence of 5 J/cm2 and t = 600 ns atdifferent distances from the target.
E. Amer et al. / Applied Surface Science 256 (2010) 4633–4641 4637
F ) t = 2a
tp
3
flsetTiseYottb1
wdbrts
ig. 6. A time series of intensity maps of Ti at a laser fluence of 5 J/cm2. (a) t = 50 ns, (bnd bright represents high transmission.
igated, since it does not influence the main features (the centralart of the plume) which we discuss in this paper.
.2. Ti results
Time series of intensity maps and phase maps of Ti at a laseruence of 5 J/cm2 are shown in Figs. 6 and 7, respectively. Fig. 6hows that at early time (t = 50 ns) the probe beam is to a largextent absorbed by the induced plume. At later times the front ofhe induced plume and the shock wave front can be distinguished.he figures show also that the dark region close to the target result-ng from the absorption of the probe beam by the induced plumemears out with time and disappears at t = 800 ns. Fig. 8 shows thextinction of the probe beam (532 nm) taken as an average from= 1.42 mm to Y = 1.99 mm for different time delays as a functionf Z. The peak appearing towards higher Z-values is the position ofhe shock wave front for different times. It is seen that the extinc-ion of the laser beam decreases with time. About 50% of the probeeam is absorbed at t = 50 ns which decreases with time to be about0% at t = 800 ns.
The phase maps in Fig. 7 show that the induced plume and shockave expand homogenously with time. The dark regions within the
isturbed region are a result from a lower refractive index and theright regions close to the shock front are a result from a higherefractive index as compared to undisturbed air. It is also seen thathe phase maps are almost symmetrical wherefore Radon inver-ion can be used to estimate the 3D refractive index distribution.
00 ns, (c) t = 400 ns, (d) t = 600 ns and (e) t = 800 ns. Dark represents low transmission
The refractive index profiles at Z = 0.12 mm and at 0.25 mm fromthe centre at t = 200 ns and t = 800 ns are shown in Fig. 9. The fig-ure shows that the refractive index at early time is lower than one,indicating the presence of plasma [19,26]. At later time the refrac-tive index is higher than one, indicating the presence of a neutralgas. The electron number density has been calculated from thereconstructed refractive index and it is found to be of the orderof 1018 cm−3. The procedure of the electron number density calcu-lation is described in [26].
4. Discussion
A comparison of intensity maps and phase maps of Zn and Ti at alaser fluence of 5 J/cm2 and a time delay of 400 ns is shown in Fig. 10.Fig. 10(a) shows a comparison of the intensity maps (Ti upper partand Zn lower part). Fig. 10(b) shows a comparison of the phasemaps (Ti upper part and Zn lower part). The figures show that theplume structure is different for Zn and Ti for the same laser settingsand surrounding gas. The comparison of the intensity maps showsthat in the case of Zn streaks appear as dark and bright lines closeto each other indicating deviation of the probe beam. In contrast,the intensity map of Ti shows a dark region close to the target that
indicates absorption of the probe laser beam by the induced plume.From the phase maps, in case of Zn, bright streaks emerging fromthe surface can be seen while these streaks are not seen in the caseof Ti. Instead a homogenous phase map can be observed. In additionit is seen that the radius of the shock wave is about 16% larger for
4638 E. Amer et al. / Applied Surface Science 256 (2010) 4633–4641
2. (a)
taa
astp
Ffl
Fig. 7. A time series of phase maps of Ti at a laser fluence of 5 J/cm
he Zn target as compared to the Ti target, which indicates thatbout 2.1 times more energy has been released from the Zn targetssuming the validity of the point explosion model [27].
Fig. 11 shows the observed patterns in the intensity maps for Znnd Ti at t = 600 ns and laser fluence of 5 J/cm2 (lower part) and aketch representing possible physical mechanisms that may causehe differences in these patterns (upper part). For a comparisonurpose the coordinate system is mirrored in case of Zn. From the
ig. 8. The extinction of the probe laser beam (532 nm) by the Ti plume at a laseruence of 5 J/cm2 at different time delays, average from Y = 1.42 mm to Y = 1.99 mm.
t = 50 ns, (b) t = 200 ns, (c) t = 400 ns, (d) t = 600 ns and (e) t = 800 ns.
intensity maps, the following regions with different patterns havebeen distinguished:
˛ Undisturbed air.
ˇ Main shock wave.� Compressed air.ı Streaks in case of Zn.� Absorption in case of Ti.
Fig. 9. Refractive index profiles at Z = 0.12 mm and at 0.25 mm from the centre forTi at a laser fluence of 5 J/cm2 at t = 200 ns and t = 800 ns.
E. Amer et al. / Applied Surface Science 256 (2010) 4633–4641 4639
F nce op
P
A
D
G
Fp
ig. 10. A comparison of intensity maps and phase maps of Zn and Ti at a laser flueart) and (b) phase maps (Ti upper part and Zn lower part).
ossible physical mechanisms that cause the previous patterns are:
Undisturbed air (no spatial changes).B Pressure jump at the shock front.C Compressed air with high density.
Vapor with particle size in the order of nm (Rayleigh scattering).E Large particles (ejected drops or condensed particles) in the order
of �m (Mie scattering).F Plasma, inverse bremsstrahlung and photoionization absorption.
Line absorption.
ig. 11. Illustration of different observed patterns (intensity map, lower part) andossible physical mechanisms (upper part) for Zn and Ti.
f 5 J/cm2 and a time delay of 400 ns. (a) Intensity maps (Ti upper part and Zn lower
The physical properties for Zn and Ti targets are presented inTable 1. The thermodynamic critical temperature has been takenfrom Martynyuk [28]. The surface reflectivity for normal incidenceand the absorption coefficient at wavelength of 1064 nm weretaken from Lide [29]. The values given are for room temperatureand therefore only an indicator. In particular the reflectivity canstrongly depend on temperature but also on the original surfaceconditions, that are oxidized, although polished. Some of the prop-erties determine the process in the solid material and thus onlyindirectly affect on the expanding plume, while the last two prop-erties have a direct effect on the plume expansion.
From the values in Table 1, it is seen that the melting point, boil-ing point, latent heat and the thermodynamic critical temperatureof Zn are much lower than these of Ti.
Based on the differences in the plume structure indicated inFig. 10 and the differences in the physical properties tabled inTable 1, a discussion about possible causes of these differences canbe performed. In the case of Zn, the streaks that are visible in boththe intensity maps and the phase maps show the paths of the mate-rial emerging from the surface as a result of the ablation process.The results show that at the positions where these streaks appear(from the intensity maps) the probe beam has deviated and (from
the phase maps) the refractive index is larger than the undisturbedrefractive index. Hence there is a refractive index gradient at thepositions of the streaks that causes the probe beam to deviate. Thisrefractive index gradient is a result of the concentration gradient of
Table 1Physical properties of Zn and Ti.
Zn Ti
Density (kg/m3) 7140 4500Melting point (◦C) 419.53 1670Boiling point (◦C) 907 3287Thermodynamic critical temperature (◦C) 2657 7617Latent heat of fusion (J/g) 113 390Latent heat of evaporation (J/g) 1748 8893Thermal conductivity (W m−1 K−1) 116 21.9Surface reflectivity at 1064 nm 0.684 0.55Absorption coefficient (m−1) at 1064 nm 4.7 × 107 3.9 × 107
Atomic weight (g/mole) 65.409 47.867First ionization energy (eV) 9.39 6.83
4 ce Scie
twldptitncsnooahsotaCialatmicvabetoc(ifasm
gtadrpotiiti2pirbospric
640 E. Amer et al. / Applied Surfa
he ablated Zn material. Since these streaks have certain directionse may conclude that the streaks consist of ablated material with
arge momentum in the direction normal to the target. So far noistinct criterion could be found that associates the observed localattern character to mechanism (D) or (E) and corresponding par-icle size, respectively. Thus, in the following we discuss indirectndications. One likely hypothesis is that large scale particles (inhe order of �m) appear as a result of a phase explosion mecha-ism. The thermal properties of Zn show that its thermodynamicritical temperature is very low (2657 ◦C) and can be reached easilyo that phase explosion can take place. The phase explosion mecha-ism has been observed by several authors as a result of the impactf a nanosecond laser pulse on metallic targets. Xu and Willis [16]bserved the phase explosion using a KrF excimer laser (� = 248 nmnd pulse duration of 25 ns) on a nickel target at a laser fluenceigher than 5 J/cm2. Fishburn et al. [17] observed the phase explo-ion of 30 ns pulse duration at a laser fluence higher than 7 J/cm2
n an aluminium target, while Porneala and Willis [18] reportedhat the phase explosion was observed at 5.2 J/cm2 for aluminiumblated by a Nd:YAG laser (� = 1064 nm and pulse duration of 5 ns).ommon for these observations is that there appears to be a jump
n the ablation rate at a point where the phase explosion mech-nism starts to appear, which could be a possible cause for thearger amount of released energy from the Zn target. The streaksre not seen in either the intensity map or phase map at earlierime, they start to appear at t = 200 ns. Possibly, at earlier time nor-
al vaporization and plasma formation takes place. This is seenn Fig. 2(a), a bulk dark region close to the target due to almostomplete absorption of the probe laser beam. Time is needed forapor bubbles to grow to the critical size in the superheated liquidnd for the phase explosion to take place. This time seems to beetween 50 ns and 200 ns in the present case. A different hypoth-sis is that these large scale particles are due to condensation ofhe Zn vapor. The time needed for the condensation is in the orderf �s and also if this condensation takes place it should happenlose to the contact front between the plume and compressed airaway from the target) where significant cooling occurs [30]. Thiss not seen in the phase difference profiles since higher phase dif-erence is observed close to the target and decrease with distancelong the streaks (see Fig. 5). As a result we suggest that these largecale particles emerge from the surface due to the phase explosionechanism.In the case of Ti, the dark bulk region shown close to the tar-
et in the intensity map at early times smears out with increasingime and completely disappears at later times, indicating that thebsorption of the probe beam is significant at earlier times andecreases with time. The 3D refractive index results show that theefractive index at earlier times is lower than one, indicating theresence of plasma. At later time the refractive index is higher thanne, indicating the presence of a neutral gas. This could be due tohe recombination of electrons and ions with time. Normal vapor-zation mechanism takes place in the case of a Ti target and thenduced vapor is ionized due to the interaction with the rest ofhe pumping laser beam (1064 nm) during its 12 ns duration. Themages show that afterwards the Ti-plasma is still sustained until00 ns and recombined with time. The induced plasma absorbs therobe laser beam by inverse bremsstrahlung (IB) and photoion-
zation (PI) at early time and with increasing the time the plasmaecombines due to cooling and hence the absorption is reduced,eing consistent with the intensity maps. IB involves absorptionf photons by free electrons which are accelerated during colli-
ion with neutral or ionized atoms. Photon absorption also can takelace by PI process; in this case the photon energy should be compa-able to the ionization energy of excited atoms which are producedn the plume by electron-atoms collision processes. In the presentase, the line absorption (G) does not contribute, as no spectral lines
nce 256 (2010) 4633–4641
for Ti coincide with the wavelength of the probing beam wave-length of 532 nm (although a few lines are located closely in thespectrum, however being of narrow bandwidth).
Based on the discussion above a qualitative image of the mech-anisms involved in the formation of the results comes into mind.In the case of Zn, the normal evaporation mechanism takes placeat earlier time and then after a certain time when the phaseexplosion takes place a mixture of gas and liquid droplets arethrown out from the surface that mix the surrounding gas (air)and result in a non-homogeneous plume. This second mecha-nism does not happen for the Ti target at this low laser fluencewherefore the generated plume expands homogenously withtime.
5. Conclusion
• Pulsed digital holographic interferometry allows calculating timeresolved intensity maps and phase maps providing informationabout the transmission of the probing laser beam and the refrac-tive index change along its path.
• Zn and Ti have different plume structures at the same lasersettings and surrounding gas, depending on their thermal andoptical properties. Zn shows streaks, while Ti shows a dark bulkregion close to the surface.
• For Zn a higher refractive index at the streak locations indicateshigher density due to off-streaming material.
• For Ti the reconstructed refractive index field shows that therefractive index is lower than one (n < 1) close to the target atearly times as evidence for the presence of plasma and in turnabsorption, as confirmed by the intensity map.
• The amount of released energy in case of Zn is about 2.1 times thatreleased in case of Ti assuming the validity of the point explosionmodel.
• At a laser fluence of 5 J/cm2, phase explosion appears to be theablation mechanism in case of Zn, while for Ti normal vaporiza-tion seems to be the dominant mechanism.
Acknowledgements
We would like to acknowledge the Egyptian government andthe Kempe Foundation for the financial support of Eynas Amer.
References
[1] H. Li, S. Costil, V. Barnier, R. Oltra, O. Heintz, C. Coddet, Surface modificationsinduced by nanosecond pulsed Nd:YAG laser irradiation of metallic substrates,Surf. Coat. Technol. 201 (2006) 1383–1392.
[2] L.L. Sartinska, S. Barchikovski, N. Wagenda, B.M. Rud, I.I. Timofeeva, Laserinduced modification of surface structures, Appl. Surf. Sci. 253 (2007)4295–4299.
[3] M.A. Shannon, X.L. Mao, A. Fernandez, W.-T. Chan, R.E. Russo, Laser ablationmass removal versus incident power density during solid sampling for induc-tively coupled plasma atomic emission spectroscopy, Anal. Chem. 67 (1995)4522–4529.
[4] N. Miroshnikova, M. Sjödahl, P. Gren, I. Sarady, Percussion hole drilling of met-als with a fourth-harmonic Nd:YAG laser studied by defocused laser specklecorrelation, Appl. Opt. 44 (2005) 3403–3408.
[5] R. Pini, S. Siano, R. Salimbeni, M. Pasquinucci, M. Miccio, Tests of laser cleaningon archeological metal artefacts, J. Cult. Heritage 1 (2000) S129–S137.
[6] S. Weimantel, G. Reie, Pulsed laser deposition of adherent hexagonal/cubicboron nitride layer systems at high growth rates, Appl. Surf. Sci. 197–198 (2002)331–337.
[7] D. Bleiner, Z.Y. Chen, D. Autrique, A. Bogaerts, Role of laser-induced melt-ing and vaporization of metals during ICP-MS and LIBS analysis, investigatedwith computer simulations and experiments, J. Anal. At. Spectrom. 21 (2006)
910–921.
[8] M Hauer, D.J. Funk, T. Lippert, A. Wokaun, Time resolved study of the laserablation induced shockwave, Thin Solid Films 453 (2004) 584–588.
[9] X. Zeng, X.L. Mao, R. Greif, R.E. Russo, Experimental investigation of ablationefficiency and plasma expansion during femtosecond and nanosecond laserablation of silicon, Appl. Phys. A: Mater. Sci. Process. 80 (2005) 237–241.
ce Scie
[
[
[
[[
[
[
[
[
[
[
[
[
[
[
[[
[
[
E. Amer et al. / Applied Surfa
10] E. Amer, P. Gren, A.F.H. Kaplan, M. Sjödahl, Impact of an extended source inlaser ablation using pulsed digital holographic interferometry and modelling,Appl. Surf. Sci. 255 (2009) 8917–8925.
11] S.B. Wen, X.L. Mao, R. Greif, R.E. Russo, Laser ablation induced vapor plumeexpansion into a background gas. II. Experimental analysis, J. Appl. Phys. 101(2007) 023115.
12] J.F.Y. Gravel, D. Boudreau, Study by focused shadowgraphy of the effect oflaser irradiance on laser-induced plasma formation and ablation rate in variousgases, Spectrochim. Acta B-At. Spectrosc. 64 (2009) 56–66.
13] C. Phipps, Laser Ablation and its Applications, Springer, New York, USA, 2007.14] A. Miotello, R. Kelly, Critical-assessment of thermal models for laser sputtering
at high fluences, Appl. Phys. Lett. 67 (1995) 3535–3537.15] N.M. Bulgakova, A.V. Bulgakov, Pulsed laser ablation of solids: transition from
normal vaporization to phase explosion, Appl. Phys. A: Mater. Sci. Process. 73(2001) 199–208.
16] X.F. Xu, D.A. Willis, Non-equilibrium phase change in metal induced bynanosecond pulsed laser irradiation, J. Heat Transf.-Trans. ASME 124 (2002)293–298.
17] J.M. Fishburn, M.J. Withford, D.W. Coutts, J.A. Piper, Study of the fluence depen-dent interplay between laser induced material removal mechanisms in metals:vaporization, melt displacement and melt ejection, Appl. Surf. Sci. 252 (2006)
5182–5188.
18] C. Porneala, D.A. Willis, Observation of nanosecond laser-induced phase explo-sion in aluminum, Appl. Phys. Lett. 89 (2006) 211121.
19] H. Schittenhelm, G. Callies, P. Berger, H. Hugel, Time-resolved interferometricinvestigations of the KrF-laser-induced interaction zone, Appl. Surf. Sci. 110(1997) 493–497.
[
[
nce 256 (2010) 4633–4641 4641
20] G. Callies, P. Berger, H. Hugel, Time-resolved observation of gas-dynamic dis-continuities arising during excimer-laser ablation and their interpretation, J.Phys. D: Appl. Phys. 28 (1995) 794–806.
21] S.H. Jeong, R. Greif, R.E. Russo, Shock wave and material vapour plume prop-agation during excimer laser ablation of aluminium samples, J. Phys. D: Appl.Phys. 32 (1999) 2578–2585.
22] T. Kreis, Holographic Interferometry Principles and Methods, Akademie Ferlag,Berlin, 1996.
23] M. Takeda, H. Ina, S. Kobayashi, Fourier-transform method of fringe-patternanalysis for computer-based topography and interferometry, J. Opt. Soc. Am.72 (1982) 156–160.
24] P. Gren, S. Schedin, X. Li, Tomographic reconstruction of transient acoustic fieldsrecorded by pulsed TV holography, Appl. Opt. 37 (1998) 834–840.
25] C.M. Vest, Holographic Interferometry, John Wiley, New York, 1979.26] E. Amer, P. Gren, M. Sjödahl, Laser-ablation-induced refractive index fields
studied using pulsed digital holographic interferometry, Opt. Laser Eng. 47(2009) 793–799.
27] E. Amer, P. Gren, M. Sjödahl, Shock wave generation in laser ablation studiedusing pulsed digital holographic interferometry, J. Phys. D: Appl. Phys. 41 (2008)215502.
28] M.M. Martynyuk, Critical constants of metals, Russ. J. Phys. Chem. 57 (1983)
494–500.
29] D.R. Lide, CRC Handbook of Chemistry and Physics: A Ready-Reference Book ofChemical and Physical Data, 85th ed., CRC Press, Boca Raton, Florida, 2004.
30] S.B. Wen, X.L. Mao, R. Greif, R.E. Russo, Experimental and theoretical studiesof particle generation after laser ablation of copper with a background gas atatmospheric pressure, J. Appl. Phys. 101 (2007) 123105.
![Gas-phase growth of diameter-controlled carbon nanotubes · precursor, Vander Wal et al. [10] employed a pulsed laser ablation (PLA) to preform metal catalysts particles so that the](https://static.documents.pub/doc/80x56/5f5b90ed3a983439ba0910a6/gas-phase-growth-of-diameter-controlled-carbon-nanotubes-precursor-vander-wal-et.jpg)
