SEM Analysis

SEM analysis reveals the surface modification of the zinc and illustrates the formation of craters for all fluences. However, an image of Figure 1a is an example which reveals the overall modified surface of zinc (crater and ripple — formation) after ablation at a laser irradiance of 100 GW/cm² corresponding to pulse energy of 200 mJ. The whole crater is displayed (showing the crater + part of un-ablated surface) in Figure 1a. Whereas Figure 1b exhibits the magnified view of the central ablated area of zinc (inset of Fig. 1a) in which only ripples are seen.

Fig. 1. SEM images revealing the surface modification of zinc at laser irradiance of 100 GW/cm². (a) The overall modified surface of zinc after laser ablation with crater and ripple- formation. (b) The magnified view of the central ablated area with only an appearance of ripples.

SEM images of Figure 2 illustrate the effect of laser irradiance on the surface morphology of the irradiated zinc targets. Figure 2a–2d represents the central ablated region. It is observed that laser irradiated surfaces are corrugated and rippled at central ablated areas. Figure 2a shows the growth of indistinct ripples for the lowest irradiance of 25 GW/cm². When the irradiance increases up to 50 GW/cm², distinct and elevated ripples are seen in Figure 2b. Their stability and smoothness also decreases. The protruding surface morphology shows sinusoidal behavior along with bead like structures. When the irradiance increases up to 75 GW/cm² less distinct and narrow ripples are seen in Figure 2c. The periodicity does not change significantly. Another feature between the ripples is the remnants of turbulent melt flow. The regularity and periodicity of the ripples is enhanced at the highest irradiance of 100 GW/cm² shown in Figure 2d. Their appearance becomes less distinct. Interference between the incident beam and the scattered wave is one of the proposed mechanism for the formation of these ripples (Sipe et al., Reference Sipe, Young, Preston and Van Driel1983). The average periodicity of the ripples varies from 20 to 30 µm but it is much greater than the wavelength of the incident laser beam (1.064 µm) for all irradiances. This suggests the exclusion of contribution of interference effects for the formation of these rippled structures. Certain hydrodynamical (instabilities) mechanisms may be more reasonably accounted for the growth of such kind of structures (Liu et al., Reference Liu, Jiang, Yang, Guan and Dai2011). Hydrodynamic expulsion is controlled by a competition between the inertial forces induced by the pressure field into the viscous melt and the surface tension that tends to maintain cohesion (Bleiner et al., Reference Bleiner and Bogaerts2006). These large scale ripple patterns look more like an unstable interface between the pool water and the wind blowing along its surface. The melted surface and expanding plasma plume above the target surface play the role of water and wind, respectively (Liu et al., Reference Liu, Jiang, Yang, Guan and Dai2011). These structures can also be considered as frozen morphology due to capillary waves (Young et al., Reference Young, Sipe and Van Driel1984). The elevation of the ripples increases from 25 GW/cm² to 50 GW/cm². However, further increase in laser irradiance up to 100 GW/cm² causes reduction in the elevation and clarity of the ripples. The initial increase is attributed to the enhanced energy deposition. When the laser irradiance increases above 50 GW/cm² ripples become less distinct. This is explainable on the basis of shielding effect of plasma (Farid et al., Reference Farid, Bashir and Khaliq2012). When the energy deposition increases plume temperature and density also increases. This plume with species of high kinetic energies and high fluxes will be able to absorb incoming laser radiations and therefore less laser energy will be deposited on the zinc target. This shielding effect will be responsible for reducing the electron temperature and number density of zinc plasma as has been confirmed by the LIBS analysis (Section 2).

Fig. 2. SEM images of the central ablated region of zinc targets at various laser irradiances of (a) 25, (b) 50, (c) 75, and (d) 100 GW/cm².

SEM images of Figures 3a–3d present the inner boundaries of the laser ablated zinc. Cones and wave-like ridges are dominant structural features. Forward peaked cones with wide base and spherical top are seen at the surface irradiated with laser irradiance of 25 GW/cm² as shown in Figure 3a. Preferential ablation due to evaporation resistive impurities, defects and hydrodynamic sputtering may be accounted for their formation (Kawakami et al., Reference Kawakami and Ozawa2003; Dolgaev et al., Reference Dolgaev, Fernández-Pradas, Morenza, Serra and Shafeev2006). Increasing the irradiance up to 50 GW/cm² conical features are almost vanished and ridge formation is enhanced as shown in Figure 3b. Flakes are also formed in the inter ridged regions indicating the exfoliational sputtering. Further increase in irradiance up to 75 GW/cm² increases the melted imprints in form of uplifted ridges and wavy patterns as shown in Figure 3c. Enhanced recoil pressure of the plasma splashes more material toward the boundary and causes ridge formation (Ko et al., Reference Ko, Pan, Hwang, Chung, Ryu and Costas2007). The melt displacement and expulsion become do minanat (Cristoforetti et al., Reference Cristoforetti, Legnaioli, Palleschi, Tognoni and Benedetti2008). Melt pool perturbation may also lead to the growth of Rayleigh-Taylor instability resulting in the formation of these wavy patterns (Willis et al., Reference Willis and Xu2000). When the irradiance increases to the 100 GW/cm², cones with increased diameter, thick underlying ridges, and deep valleys are formed as shown in Figure 3d. Laser energy deposition at the boundary is smaller as compared to central ablated area because of the Gaussian distribution of the laser pulse. Shielding effects are therefore not dominant at peripheries due to lower irradiances (Cabalín et al., Reference Cabalín, Romero, Baena and Laserna1999; Bleiner, Reference Bleiner and Bogaerts2006). Hence, enhanced energy deposition with the increasing laser irradiance is responsible for the generation of these large size cones. The hydrodynamic pressure of the expanding plasma drives a shock wave toward the target surface in addition to the heating. The combined effect of heat and shock wave causes melting, evaporation, and sputtering of the target surface (Torrisi et al., Reference Torrisi, Gammino, Mezzasalma, Gentile, Krása, Láska, Rohlena, Badziak, Parys and Wolowski2003). SEM images of Figures 4a–4d show the outer boundary region of ablated zinc targets. It comprises of the displaced material from the center and re-deposited plasma condensates in addition to the direct laser heating effects. The structures observed at outer boundary are cones, flakes, and cavities. The heating of subsurface layer and rapid volume expansion of absorbed gases in the underlying pores of the target surface lead to the formation of these cavities (Lappalainen et al., Reference Lappalainen, Frantti and Lantto1998). Explosive relaxation of the mechanical stresses in the superficial layer may also be accounted for the formation of these cavities (Tokarev et al., Reference Tokarev2006). The size and number of cavities increase with the increase of laser irradiance up to a certain value and then decrease. It may be due to their re-filling with the melt.

Fig. 3. SEM images of the inner boundaries of the ablated zinc targets at various laser irradiances of (a) 25, (b) 50, (c) 75, and (d) 100 GW/cm².

Fig. 4. SEM images of the outer boundaries of the ablated zinc targets at various laser irradiances of (a) 25, (b) 50, (c) 75, and (d) 100 GW/cm².

LIBS Analysis

In order to correlate the ablation mechanism of irradiated zinc target with the plasma parameters, LIBS analysis was also performed. The plasma is characterized by its emission intensity, temperature, and number density of charged particles. All of these parameters are evaluated at various laser irradiances. Figure 5 shows the laser induced breakdown spectrum of zinc plasma obtained with 63 GW/cm² irradiance in argon ambient at a pressure of 20 Torr. It contains different spectral lines from neutral zinc atoms at 334.5 nm, 468.0 nm, 472.2 nm, 481.0 nm, and 636.2 nm and from singly ionized zinc ions at 491.16 nm, 492.40 nm (shown in the inset) wavelengths. The rest of the lines correspond to Ar(I) at 696.54 nm, 763.51 nm, 810.37 nm, 811.53 nm, etc., and indicate its breakdown down (Shaikh et al., Reference Shaikh, Rashid, Hafeez, Jamil and Baig2006a). The obtained plasma is a mixture of ablated metal vapor and background gas.

Fig. 5. The emission spectrum of laser induced zinc plasma under the ambient environment of Argon at a pressure of 20 Torr by using Nd:YAG laser (λ = 1064 nm, t =10 ns) at irradiance of 63 GW/cm² .

Figure 6 shows the variation of emission intensity with the laser irradiance for the selected spectral lines corresponding to neutral zinc atoms. It depicts an initial increase of emission intensities with the increasing laser irradiance reaches a maximum then decreases followed by almost constant trend.

Fig. 6. (Color online) The variation of emission intensity of spectral lines of neutral zinc atoms at various laser irradiances ranging from 13 GW/cm² to 100 GW/cm².

The obtained emission spectra are used to calculate the excitation temperature and number density of the zinc plasma. Few assumptions are made prior to the calculations which generally hold good for the characteristics of the laser induced plasma in the concerned irradiance range. The plasma is considered to be in local thermal dynamical equilibrium (LTE) due to the dominance of collisional processes among the species and hence characterized by a homogenous temperature. The population in the excited states follows the Boltzmann distribution given by the following equation (Bashir et al., Reference Bashir, Farid, Mahmood and Rafique2012a; Harilal et al., Reference Harilal, Bindhu, Nampoori and Vallabhan1998; Iida et al., Reference Iida1990; Cristoforetti et al., Reference Cristoforetti, Palleschi, Salvetti and Tognoni2004; Hermann et al., Reference Hermann, Boulmer-Leborgne, Dubreuil and Mihailescu1993).

(1)$$ln\lpar {\rm \lambda}_{\rm mn}I_{\rm mn}/g_{\rm m}A_{\rm mn}\rpar =-E_{\rm m}/KT_{e}+ln\lpar N\lpar T\rpar /U\lpar T\rpar \rpar \comma \;$$

where λ_mn, I _mn, g _m, and A _mn are wavelength, intensity, statistical weight, and transition probability of the upper level state m, respectively. On the right-hand side U(T), N(T), K, E _m, and T _e are partition function, total number density, Boltzmann constant, energy of the upper level state, and electron temperature, respectively. Boltzmann plot of the logarithmic term on the left-hand side versus the E _m gives a straight line whose slope is equal to the −1/KT _e under the assumption made that the distribution is Boltzmann. The relevant spectroscopic parameters for the transitions of the zinc plasma used in the calculation are listed in Table 1 and have been taken from Shaikh et al. (Reference Shaikh, Rashid, Hafeez, Jamil and Baig2006; NIST database).

Table 1. The spectroscopic data for the selected lines used to calculate the electron temperature and number density of laser induced zinc plasma (Sansonetti et al., 2005)

The transition lines at 334.5 nm (4s4d ³D₃ → 4s4p ³P₂), 468.0 nm (4s5s ³S₁ → 4s4p ³P₀), 472.2 nm (4s5s ³S₁ → 4s4p ³P₁), 481.0(4s5s ³S₁ → 4s4p ³P₂), and 636.2(4s4d ¹D₂ → 4s4p ¹P₁) are used for the calculation of electron temperature.

The variation of electron temperature with laser irradiance is shown in Figure 7. It increases slightly from 6126 K to 6240 K with increasing laser irradiance from 25 GW/cm² to 63 GW/cm². With further increase in laser irradiance from 63 GW/cm² to 75 GW/cm² the electron temperature drops down from 6240 K to 6030 K. With further increasing fluence up to a maximum value of 100 GW/cm², the saturation with a negligible change (within error bar) in the electron temperature is achieved.

Fig. 7. The variation of electron temperature of laser induced zinc plasma obtained at various laser irradiances ranging from 13 GW/cm² to 100 GW/cm².

The plasma number density has been extracted from the profile of the line spectra obtained. Plasma species are under the influence of electric field of fast moving electron and slow ions. The perturbing electric field shifts the energy levels of the species leading to broadening called the Stark broadening. Other broadening mechanisms like Doppler and pressure broadening contribute little in this case and hence can be ignored. The relation of the full width at half maximum (FWHM) (Δλ_1/2) of the stark broadened profile of the spectral lines and the electron number density of neon is given by the equation (Bashir et al., Reference Bashir, Farid, Mahmood and Rafique2012a; Harilal et al., Reference Harilal, Bindhu, Nampoori and Vallabhan1998; Iida et al., Reference Iida1990; Cristoforetti et al., Reference Cristoforetti, Palleschi, Salvetti and Tognoni2004; Hermann et al., Reference Hermann, Boulmer-Leborgne, Dubreuil and Mihailescu1993).

(2)$$N_{e}=\lpar \Delta {\rm \lambda}_{1/2}/2w\rpar \times 10^{16} {\rm cm}^{-3}.$$

Where w is the impact width parameter. The Zn(I) line at 636.23 is used to calculate the electron number density.

The self-absorption for the spectral line 636.23 nm has been considered. The selected line must be sharp and the least broaden for fulfilling the criteria of minimum self-absorption. The selected line to be used for the calculations of number density is 632.23 nm for Zn(I) and it is sharp and the least broaden line as compared to all other lines of the obtained spectra. In order to check the self-absorption (SA) of a spectral line, following relation can be utilized for the evaluation of the parameter of SA (Cristoforetti et al., Reference Cristoforetti, Lorenzetti, Legnaioli and Palleschi2010b):

(3)$${\rm SA}=\lpar \Delta {\rm \lambda} /2w_{s}N_{e}\rpar ^{1/ {\rm \alpha}}.$$

Where α = –0.54. The Δλ, w _s, and N _e is experimentally determined stark width, half width stark broadening parameter, and electron number density calculated from the H _α line as (Sherbini et al., Reference Al-Sherbini, Al-Sherbini, Hegazy, Cristoforetti, Legnaioli, Palleschi, Pardini, Salvetti and Tognoni2005):

(4)$$n_{e}\lpar {\rm cm}^{-3}\rpar =C\lpar {\rm \lambda}\comma \; T\rpar \lpar \Delta {\rm \lambda}\rpar ^{3/2}\comma \;$$

where Δλ is FWHM of H _α line and C(λ, T) is a coefficient weakly dependent on density and temperature and has been tabulated in Griem et al. (Reference Griem1964). The value for Δλ is obtained by fitting the line with a Voigt profile and by considering the width of its Lorentzian component (Cristoforetti et al., Reference Cristoforetti, Lorenzetti, Legnaioli and Palleschi2010b). The Stark broadening parameter for 636.23 nm Zn(I) line has been taken from the work reported in Sherbini et al. (Reference El Sherbini, Aboulfotouh, Rashid, Allam, Al-Kaoud, Dakrouri and El. Sherbini2013). When SA value is nearly equal to 1 and it indicates that self-absorption effects are negligible while SA≪1 indicates that nonlinear effects are strong (Cristoforetti et al., Reference Cristoforetti, Lorenzetti, Legnaioli and Palleschi2010b). By putting data from our experiment the evaluated value of SA comes out to be nearly equal to one and therefore indicates that self-absorption is negligible for our case.

The self-absorption coefficient can also be calculated as (Sherbini et al., Reference Sherbini and Saad Al Aamer2012):

(5)$${\rm SA}=I\lpar {\rm \lambda}_{0}\rpar /I_{0}\lpar {\rm \lambda}_{0}\rpar =\lpar \Delta {\rm \lambda}_{\rm l}/ \Delta {\rm \lambda}_{0}\rpar ^{1/ {\rm \alpha}}={\lpar n_{e}^{\rm l} / n_{e}^{H {\rm \alpha}}\rpar }^{1/ {\rm \alpha}}.$$

Where I(λ₀) and I ₀(λ₀) are the spectral intensity of line in limit of negligible self-absorption and experimentally measured line height respectively. Δλ₀ and Δλ_l are the Lorentzian component of spectral line if the line is optically thin and broadened FWHM of Lorentzian component of the same line because of self-absorption process respectively. n _e^l and n _e^Hα are the number densities evaluated from line under investigation and H _α line, respectively. The obtained spectral line is plotted and fitted with the Lorentizan profile to obtain the FWHM and w (impact width parameter).

The variation of electron number density with increasing laser irradiance is shown in Figure 8. It increases from 1.3586 × 10¹⁸cm⁻³ to 1.4433 × 10¹⁸cm⁻³ with increasing laser irradiance from 25 GW/cm² to 63 GW/cm². With further increase in laser irradiance from 63 GW/cm² to 75 GW/cm² it drops down slightly to a value of 1.4296 × 10¹⁸ cm⁻³. With further increasing fluence up to a maximum value of 100 GW/cm², the saturation with a negligible change (within error bar) in the electron density is achieved.

Fig. 8. The variation of electron number density of zinc plasma at various laser irradiances ranging from 13 GW/cm² to 100 GW/cm².

The observed trend in electron temperature and number density is divided into three regions: an initial increase with the increase of laser irradiance reaches a maximum at 63 GW/cm² in the first region, then decrease in the second region, and finally saturation is achieved in the third region. This can be explained as follows: when a nanosecond laser interacts with the material it heats up the target until its melting point is reached and eventually causes it to eject into the form of vapor (Farid et al., Reference Farid, Cong, Hongbei and Hongbin2013). Initially, at lower irradiances the ablation rate increases with the increase of laser irradiance which in turn increases the temperature and charge particle density in the plasma (Luo et al., Reference Luo, Zhao, Sun, Gao, Tang, Wang and Zhao2010). The background pressure also effectively confines the plasma species which further adds to increase the temperature and density of charged particle by reducing its expansion rate. The background argon ambient at a pressure of 20 Torr is seemed to be sufficient to confine the generated target vapors. Plasma reaches its maximum temperature and number density at 63 GW/cm². This vapor plume continues to absorb the laser energy through inverse-bremsstrahlung and direct photo ionization processes and start to shield the target from the laser irradiation (Harilal et al., Reference Harilal, Bindhu, Nampoori and Vallabhan1998). Further increase in the irradiance reduces energy deposition to the target due to enhanced shielding effect of the plasma which leads to decrease the electron temperature and number density. At still higher irradiances the saturation region is observed due to the formation of a self-regulating regime in addition to the shielding effect (Cristoforetti et al., Reference Cristoforetti, Legnaioli, Palleschi, Tognoni and Benedetti2008; Bleiner et al., Reference Bleiner and Bogaerts2006). In this regime, the density, temperature, and dimensions of plume adjust in such a way that plasma absorb the same amount of laser energy to maintain the self-regulating regime (Harilal et al., Reference Harilal, Bindhu, Nampoori and Vallabhan1998; Hafeez et al., Reference Hafeez, Shaikh, Rashid and Baig2008).

Cristoforetti et al. (Reference Cristoforetti, Giacomo, Dell'Aglio, Legnaioli, Tognoni, Palleschi and Omenetto2010a) has addressed the criteria for the establishment of LTE in addition to the McWhirter criterion. The McWhirter criterion is a necessary but not the sufficient condition to ensure the existence of LTE in plasma. The condition of LTE has been assessed for plasma in three states namely stationary and homogenous, transient and homogeneous, and transient and inhomogeneous. Considering our plasma as transient and homogenous, we have fulfilled the criteria for LTE by evaluating the following two conditions: (1) The McWhirter criteria n _e (cm⁻³) = 1.6 × 10¹² (T)^1/2(ΔE _nm)³ has been fulfilled (Cristoforetti, Reference Cristoforetti, Giacomo, Dell'Aglio, Legnaioli, Tognoni, Palleschi and Omenetto2010). For the excitation temperature T _e = 6240 K, the number density of neon comes out to be of the order of 10¹⁴. (2) For homogenous and transient plasma, if the time needed for the establishment of excitation and ionization equilibrium (relaxation time) is much shorter than the time for variation of thermodynamic parameters (mainly neon) then the plasma evolves through quasi-equilibrium near LTE states (Cristoforetti et al., Reference Cristoforetti, Giacomo, Dell'Aglio, Legnaioli, Tognoni, Palleschi and Omenetto2010a). The rate coefficient for the collisional excitation from ground to first excited state of the resonance series being the slowest one has been considered for the rough estimation of this relaxation time. This can be expressed as (Cristoforetti et al., Reference Cristoforetti, Giacomo, Dell'Aglio, Legnaioli, Tognoni, Palleschi and Omenetto2010a):

(6)$$\eqalign{t_{\rm rel}=1/n_{e} \lt \!{\rm \nu}\, a_{12} \gt & =6.3 \times 10^{14} \Delta E_{21}\lpar KT\rpar ^{1/2} \cr &\exp \lpar \Delta E_{21}/KT\rpar /n_{e}f_{12} \lt \!g \gt .}$$

Where t _rel is relaxation time, ΔE ₂₁ and KT are both expressed in eV. The f ₁₂ is the transition oscillator strength, <g> is effective gaunt factor and n_e is number density in cm⁻³. The calculated time (of the order of 10⁻⁹s) came out to be reasonably shorter than the typical decay time of electron temperature and number density (approximately 1 µs).

LIBS analysis can be well correlated with the SEM investigations. The initial increase in T _e and n _e with the increasing laser irradiance is responsible for the growth of distinct and well defined ripples. After obtaining the maxima, T _e and n _e decrease because of shielding effect and formation of self-regulatory region. Due to less energy deposition of electrons to the lattice the grown structures become diffusive and narrow.