We report on the instantaneous detection of the ablation rate as a function of depth during ultrafast microdrilling of metal targets. The displacement of the ablation front has been measured with a sub-wavelength resolution using an all-optical sensor based on the laser diode self-mixing interferometry. The time dependence of the laser ablation process within the depth of aluminum and stainless steel targets has been investigated to study the evolution of the material removal rate in high aspect-ratio micromachined holes.
©2011 Optical Society of America
The applications of laser ablation for high precision micromachining are continuously expanding thanks to the reduced thermal effects and minimal collateral damage achievable onto metallic targets using extremely short laser pulses [1, 2]. The physical mechanisms controlling the ablation in metals is largely influenced by the energy transfer process from the laser pulse to the material. For pulse duration longer than a few tens of picoseconds corresponding to the characteristic electron-phonon scattering in metals, the laser ablation is mainly governed by the evaporation and the throw-out of cluster-like liquid droplets via fragmentation of the laser generated melt layer. There, the thermal load to the material causes large heat-affected zones and recast layers. Pulses with duration down to a few picoseconds or even to some hundred femtoseconds lead rather to laser ablation processes which take place through phase explosion of a superheated melt layer with subsequent ejection of nano-sized particles and negligible thermal damage to the surrounding bulk material [3, 4]. Besides the pulse duration, the laser fluence, the beam polarization, and the repetition rate are also relevant [5–8]. However, heat accumulation effects during pulsed laser irradiation may still cause the generation of some amount of molten material, which is detrimental to the quality of the laser ablation [9–11]. Post-process offline analysis of laser ablated structures and damage morphology have been extensively explored to understand such mechanisms and adjust the process technology in order to improve the achievable precision . Invasive inspections of the borehole geometry have tentatively contributed to infer the formation and the time-evolution of deep laser ablated structures . Nonetheless, to characterize high aspect-ratio micro-machined structures more reliable detection techniques are required, and the physical interactions between intense light beam and non-transparent bulk materials need diagnostic tools capable of real-time response to be fully described and unambiguously interpreted. The direct investigation of crucial parameters such as the penetration depth and the removal rate all along the process time [14–16] is therefore of paramount importance to identify different ablation phases along with providing additional insight into the ablation dynamics. Recently, we developed a laser ablation sensor which is capable to measure on-line the displacement of the ablation front within the hole capillary [17, 18]. The working principle is based on optical feedback interferometry, also used for target displacement measurements [19, 20]. We employed this contactless and self-coherent detection technique to characterize the depth-resolved evolution of the ablation surface, measuring the actual value of the penetration depth with sub-micrometric resolution.
In this paper, we apply the laser ablation sensor to study the hole-microdrilling dynamics in very thin metallic substrate (i.e., thickness below 100 µm) at extremely short time-scales. We emphasize that, to the best of our knowledge, no data are today available, neither experimental nor theoretical, covering this industrially relevant thickness range. We exploited the real-time capability of the laser self-mixing diagnostics to put in evidence distinctive ablation regimes which occur depending on the incident laser fluence. A tentative interpretation in terms of both thermal and mechanical response of the material was provided. Our results, while consistent with well-known mechanisms behind the ablation process for thick targets (≥ 100 µm), display peculiar and somehow unexpected features at shallower hole depth.
2. Laser ablation sensor: experimental configuration
A prototype ytterbium-doped fiber laser amplifier emitting at a wavelength of 1064 nm was used for the ablation experiments. Steel and aluminum targets were exposed to 120-ps pulses at a repetition rate of 110 kHz. The experiments were performed in air with a fixed position of the laser focus (percussion drilling). The linear polarization of the exit beam was converted into a circular one by using a quarter-wave plate, in order to prevent anisotropic absorption inside the metal. Figure 1 shows the layout of the interferometric sensor to monitor the time-dependence of the ablation process.
Coherent external optical feedback in a semiconductor laser diode (LD) can be exploited to reveal the displacement of the ablation front, acting as an external target with respect to the laser source . The interference between the standing wave inside the cavity and a fraction of radiation back-scattered by the bottom surface of the drilled hole results in self-mixing fringes exhibited by the output of the monitor photodiode (PD). To measure in situ the instantaneous ablation rate, the laser beam of the measuring system was coaxially aligned with the fiber laser radiation using a short wave pass dichroic beamsplitter. A plano-convex lens (LF) with a focal length of 5 cm was set to have the focal plane coinciding with the hole entrance. In all experiments, the diameter of the laser spot was about 10 μm at 1/e2 level of intensity. The energy density and the ablation peak fluence were estimated according to the method described by Liu et al. . The processing time was set using a fast mechanical shutter and the effective breakthrough time was measured using an external photodiode (PDext), as fully described in .
3. Real-time penetration depth measurement
Direct measurements of the penetration depth during percussion drilling of metallic targets with ultrashort laser pulses were performed by sampling the interferometric signal out of the detector system. In the moderate feedback regime , the sawtooth-like waveform assumed by the self-mixing interferograms exhibits a fast switching each time the interferometric phase changes by 2π, thus returning the time resolved evolution of the ablation front with a sub-wavelength resolution. The system was prealigned recording the self-mixing fringes generated by target displacement along the laser axis, using a motorized linear translation stage (not shown in Fig. 1).
3.1 Effect of laser fluence on the material removal rate
Figure 2 shows representative ablation characteristics measured during hole drilling onto carbon steel (AISI 1095) targets. A train of machining laser pulses of energy ranging from 9 μJ to 40 μJ, corresponding to a fluence on the sample surface from 3.6 J/cm2 to 15.3 J/cm2 respectively, was used to drill through the metallic targets in percussion geometry. In our operating conditions, we estimated a threshold fluence of 1.2 J/cm2. The dependence of laser ablation efficiency and material removal on laser fluence was investigated. A clear evidence of the ablation onset was provided. At the breakthrough time, the back-reflected feedback from the machining zone damped to zero, causing the interferometric sensor to stop measuring. The instantaneous value of the ablation depth during the drilling time was measured by counting the fringes of the interferometric signal. Each fringe corresponds to the displacement of the ablation front at the hole bottom by λLD/2 = 0.41 µm. The average thickness of the metallic plates was measured to be 47.74 ± 0.41 μm, within 5% of the nominal plate thickness. The interferometric signal out of the all-optical sensor allowed us to follow in real-time any change of the ablation rate deep inside the hole. In Fig. 2, different ablation regimes can be observed, depending on the laser intensity.
Specifically, at fluence up to 5.3 J/cm2, the ablation process started gently during the first 10 ms from the beginning of the laser irradiation, then increased up to reaching the same ablation rate as observed at higher fluence. Such a behavior can be ascribed to the delayed onset of thermal effects originating from heat accumulation. In fact, for a pulse length of 120 ps, heating of the substrate associated with the energy transfer from the laser pulse energy cannot be neglected, especially in the case of metals with a relatively low thermal conductivity. The temperature raise of the laser irradiated volume lead to faster ablation rates mainly due to thermal material removal mechanisms. The speed of the front position is eventually decelerating again at a very late stage of the 3.6 J/cm2 curve, as a result of local energy dissipations through the hole depth for such a fluence . However, the thermal damage became even more pronounced as the incident laser fluence was increased further or the drilling process took longer, depending also on the repetition rate of the laser beam . As far as the laser fluence was enhanced, the thermal regime could be rapidly achieved. No changes of the ablation rate were observed and the drilling process was characterized by a constant material removal rate. Thus, the percussion drilling at higher fluencies featured faster dynamics of the process, minimizing the breakthrough time . The distinctive ablation regimes as observed in Fig. 2 are well known in the literature. However, they have been inferred so far by means of indirect measurements of the average ablation rate after laser drilling experiments of metal targets with a predetermined thickness [22, 23]. By exploiting the capability of our interferometric sensor, we managed to follow in real-time any dynamical changes of the ablation regime inside laser drilled high aspect-ratio microholes.
3.2 In-depth analysis of the ablation rate evolution
The development of high-aspect ratio channels (length to diameter ratio up to 10) has been monitored for percussion drilled foils of different thicknesses. In particular, the ablation rate behaviour has been investigated and compared in Fig. 3 during the formation of microholes in carbon steel targets with a measured thickness of 46.85 ± 0.39 μm (samples A, symbols) and 92.67 ± 0.58 μm (samples B, lines), respectively.
Figure 3 shows also a comparison of the results obtained for the two investigated sets of sample at different values of the laser fluence. As the pulse duration (120-ps) exceeds the thermalization time (phonon-photon coupling) of most metals, the evidence of different ablation characteristics shown in Figs. 3(a) and 3(b), has been mainly ascribed to the thermal properties of the target material.
The time progression of the ablation front exhibited very distinctive features as the crater was drilled through. A clear evidence of the microdrilling dependence by the thermal response of the material can be given by comparing the results obtained on two carbon steel targets of different thickness, as shown in Fig. 3(b). The removal rate evolutions were perfectly overlapping at a laser fluence as high as 13.7 J/cm2, independently of the thickness of the metal plates. Nonetheless, the effect of the overall target thickness on the ablation depth could be observed as the incident laser fluence was decreased. The early stage of the drilling process was slower and less efficient on a thicker sample, as shown at 12.2 J/cm2 and more evidently at 6.9 J/cm2. At a pulse fluence slightly above the ablation threshold, the heat conduction in bulk is here assumed to be the main reason for any discrepancy on the evolution of the material removal process. Furthermore, the energy density required to breakthrough was measured to be dependent on the sample thickness, as expected. Indeed, an incident laser fluence of 6.9 J/cm2 was not effective to breakthrough a thicker sample, where the ablation front was shown to suddenly stop at a depth of about 75 μm (i.e., blind hole), as pointed by the arrow in Fig. 3(b).
The direct measurement of the laser ablation rate during microdrilling was also capable to reveal different phases of the hole evolution in the depth. In particular, three process phases could be discriminated in the thicker samples. The ignition of the drilling process was characterized by a relatively low ablation rate. This was more evident at the fluence of 6.9 J/cm2, where a removal rate lower than 25 nm/pulse was observed in the first 20 μm sample depth. Then, a continuous increase of the ablation rate was measured until the hole depth was about 70 μm. At higher laser fluencies, such an increasing ablation rate regime could be achieved since the very early stage of the process. There, the thermal removal rate mechanism is believed to be rapidly established and ablation efficiencies as high as 70 nm/pulse were measured at incident laser fluence of 13.7 J/cm2. Finally, a slope inversion and a pronounced decrease of the ablation rate with depth was consistently observed in the final 20-30 μm thickness of the hole. Several factors may concur to induce a drop of the ablation efficiency, as described in literature [9–12]. Among the others, changes of the actual capillary shape, laser beam absorption through the channel depth, and particle-ignited plasma shielding could eventually lead to local reduction of the energy density. Also, complex mechanisms of heat transport away from the ablation zone along with pulse energy deposition during the formation of high aspect-ratio hole must be accounted for [16, 22]. It is worth mentioning that so far it was not possible to characterize in real-time the different stages of the drilling process both for through and blind holes. Our direct in situ detection technique of the ablation front displacement is now capable to bring more information on the laser ablation dynamics of high aspect ratio holes in non-transparent materials.
4. Effect of the material thermal properties on the ablation regimes
Percussion drilling experiments were performed by focusing a series of 120-ps pulses onto metallic targets with different thermal conductivity. The high sensitivity of the interferometric system is demonstrated to offer additional insight into the process analysis, allowing to identify the influence of the diverse thermal properties of the irradiated material on the ablation rate.
Figure 4 shows representative ablation characteristics measured during the machining process onto stainless steel (Fe/Cr18Ni10) and carbon steel (AISI 1095) targets with a thickness of 51.34 ± 0.35 μm and 46.85 ± 0.39 μm, respectively. The thermal conductivity in stainless steel [κ(@273 K) ~15 W/mK] results three times lower than carbon steel [κ(@273 K) ~52 W/mK], which may explain a faster removal dynamics. Indeed, the ultrashort pulse laser energy transfer to the bulk material generated heat, which would reside longer into the irradiate volume of metals with a lower thermal conductivity.
Furthermore, the local temperature rapidly increased by focusing the incident radiation, yielding to a more effective pulse-to-pulse accumulation mechanism in stainless steel, where the thermal ablation regime was reached faster and the removal rate was higher than carbon steel. The curves plotted in Fig. 4 also show how the ablation regimes are still visible for both materials even at a fluence of 3.6 J/cm2. In the thermally gentle regime, which elapsed after the first 10-15 μm hole depth, the ablation front displacement proceeded with the same rate, independently from the thermal properties of the material. A bifurcation of the plotted data occurred for deeper penetrations, due to the fact that the thermal ablation regime is achieved more rapidly in the case of stainless steel, thus leading to a sudden increase of the removal rate. By increasing the laser fluence, the overall ablation rate heightens up for both materials and the bifurcation took place earlier in the process.
Figures 5(a) and 5(b) show representative ablation characteristics measured by the interferometric technique on targets of carbon steel (full diamonds), stainless steel (open diamonds) and aluminum (line), respectively. The removal rate per pulse was monitored during hole drilling by percussion technique at a laser fluence of ~5.3 J/cm2, enough to drill through the metallic targets.
In all cases, the average ablation rates could be described by a monotonically increasing function, although the evolution of the laser drilling inside the capillary featured peculiar time-dependence of the ablation front in Fig. 5(a) and the removal rate in Fig. 5(b), respectively. In particular, the heat transfer mechanism near the ablation zone was heavily affected by the thermal response of the aluminum to the incident pulses. The thermal conductivity of aluminum [κ(@273 K) ~236 W/mK] is one order of magnitude higher than stainless steel [κ(@273 K) ~15 W/mK], then we would have expected a lower ablation rate in aluminum. Nevertheless, we did observe a faster removal on the aluminum target, mainly at the beginning of the process. There, the vaporization thermodynamics is known to be also affected by the material boiling temperature, which is lower in aluminum (Tvap ≈2767 K) than stainless steel (Tvap ≈3144 K), thus leading to an earlier ablation onset under the same incident fluence. Finally, ablation rates close to 9 μm/ms could be revealed during microdrilling of aluminum targets, more than four times higher than in steel targets. By investigating the instantaneous ablation rate as a function of the target physical properties we pave the way to further understanding the ablation mechanisms and fully controlling the process parameters. It should be noted that in our experimental conditions the detection system was unable to return a sawtooth-like shape of the interferometric fringes for ablation rate higher than about 140 nm/pulse, i.e. ~12.5 μm/ms.
In conclusion, experimental evidence of real-time ablation rate measurement has been provided during percussion drilling of metal targets with ultrashort (120ps) laser pulses. High aspect-ratio channel formation have been inspected by an all-optical sensing technique, which has resulted insensitive to incoherent light from laser induced plasmas. Direct monitoring the advancement of the ablation front provides useful information on the ablation rate mechanism and the way it evolves as the hole deepens inside the sample. The effect of the incident laser fluence, sample thickness and material thermal properties on the ablation efficiency has been accurately investigated. At a relatively low laser fluence, various ablation regimes have been observed within the same process. In the early stage of the hole formation a gentle material removal mechanism is established until, owing to a heat accumulation effect, a thermal ablation regime is achieved, characterized by a faster rate. Indeed a thermal load associated with the energy transfer from the laser pulse to the bulk material cannot be avoided for 120-ps laser pulses, where the laser pulse duration is comparable to the electron-phonon coupling time. For this reason, the thermal ablation regime is reached more rapidly at higher fluencies and/or on metals with a lower thermal conductivity. In the case of thicker samples and higher aspect-ratio holes, in the last stage of the drilling process a decrease of the ablation rate has been reported, that may be ascribed to several effects like laser beam absorption through the channel depth or particle-ignited plasma shielding, etc. The onset of such perturbing effects and any changes of the ablation efficiency during the process have been directly monitored and quantified by exploiting the potential of our newly developed measuring technique based on laser self-mixing interferometry. This technique is applicable even in case of blind holes, where other similar studies (mostly based of the estimation of the average ablation rate through the measurement of the breakthrough time of a predetermined sample thickness) were unable to give any information. The assessment of crucial physical parameters such as the hole penetration depth and the laser ablation rate is demonstrated to have important implication for in situ characterization of ultrafast laser ablation. A new route is therefore accessible towards optimizing the processing parameters and developing on-line control methods to enhance the efficiency of laser ablation micromachining.
The authors acknowledge partial financial support from Regione Puglia – Project DM01.1 related with the Apulian Technological District on Mechatronics – MEDIS, and Project PS_046. The author FPM acknowledges his research contract funded by the Apulian Regional Network TRASFORMA.
References and links
1. D. Breitling, C. Föhl, F. Dausinger, T. Kononenko, and V. Konov, in Femtosecond Technology for Technical and Medical Applications, F. Dausinger, F. Lichtner and H. Lubatschowski, eds. (Springer, Berlin, 2004) Chap. 7,11.
2. C. Dorman and M. Schulze, “Picosecond micromachining update,” Laser Technik J. 5(4), 44–47 (2008). [CrossRef]
3. J. P. Colombier, P. Combis, A. Rosenfeld, I. V. Hertel, E. Audouard, and R. Stoian, “Optimized energy coupling at ultrafast laser-irradiated metal surfaces by tailoring intensity envelopes: Consequences for material removal from Al samples,” Phys. Rev. B 74(22), 224106 (2006). [CrossRef]
4. P. Lorazo, L. J. Lewis, and M. Meunier, “Thermodynamic pathways to melting, ablation, and solidification in absorbing solids under pulsed laser irradiatiuon,” Phys. Rev. B 73(13), 134108 (2006). [CrossRef]
5. S. Bruneau, J. Hermann, G. Dumitru, M. Sentis, and E. Axente, “Ultra-fast laser ablation applied to deep-drilling of metals,” Appl. Surf. Sci. 248(1-4), 299–303 (2005). [CrossRef]
6. E. G. Gamaly, N. R. Madsen, M. Duering, A. V. Rode, V. Z. Kolev, and B. Luther-Davies, “Ablation of metals with picosecond laser pulses: Evidence of long-lived nonequilibrium conditions at the surface,” Phys. Rev. B 71(17), 174405 (2005). [CrossRef]
7. M. Kraus, S. Collmer, S. Sommer, and F. Dausinger, “Microdrilling in steel with frequency-doubled ultrashort pulsed laser radiation,” JLMN-Journal of Laser Micro/Nanoengineering 3(3), 129–134 (2008).
8. M. Kraus, M. A. Ahmed, A. Michalowski, A. Voss, R. Weber, and T. Graf, “Microdrilling in steel using ultrashort pulsed laser beams with radial and azimuthal polarization,” Opt. Express 18(21), 22305–22313 (2010). [CrossRef] [PubMed]
9. P. T. Mannion, J. Magee, E. Coyne, G. M. O’Connor, and T. J. Glynn, “The effect of damage accumulation behaviour on ablation thresholds and damage morphology in ultrafast laser micro-machining of common metals in air,” Appl. Surf. Sci. 233(1-4), 275–287 (2004). [CrossRef]
10. A. Ancona, F. Röser, K. Rademaker, J. Limpert, S. Nolte, and A. Tünnermann, “High speed laser drilling of metals using a high repetition rate, high average power ultrafast fiber CPA system,” Opt. Express 16(12), 8958–8968 (2008). [CrossRef] [PubMed]
11. T. V. Kononenko, V. Konov, S. Garnov, S. Klimentov, and F. Dausinger, “Dynamics of deep short pulse laser drilling: ablative stages and light propagation,” Laser Phys. 11, 343–351 (2001).
12. A. Ancona, D. Nodop, J. Limpert, S. Nolte, and A. Tünnermann, “Microdrilling of metals with an inexpensive and compact ultra-short-pulse fiber amplified microchip laser,” Appl. Phys., A Mater. Sci. Process. 94(1), 19–24 (2009). [CrossRef]
13. G. Dumitru, V. Romano, H. P. Weber, M. Sentis, J. Hermann, S. Bruneau, W. Marine, H. Haefke, and Y. Gerbig, “Metallographical analysis of steel and hard metal substrates after deep-drilling with femtosecond laser pulses,” Appl. Surf. Sci. 208–209, 181–188 (2003). [CrossRef]
15. P. J. L. Webster, J. X. Z. Yu, B. Y. C. Leung, M. D. Anderson, V. X. D. Yang, and J. M. Fraser, “In situ 24 kHz coherent imaging of morphology change in laser percussion drilling,” Opt. Lett. 35(5), 646–648 (2010). [CrossRef] [PubMed]
16. C. S. Nielsen and P. Balling, “Deep drilling of metals with ultrashort laser pulses: a two stage process,” J. Appl. Phys. 99(9), 093101 (2006). [CrossRef]
17. F. P. Mezzapesa, A. Ancona, T. Sibillano, F. De Lucia, M. Dabbicco, P. M. Lugarà, and G. Scamarcio, “High-resolution monitoring of the hole depth during ultrafast laser ablation drilling by diode-laser self-mixing interferometry,” Opt. Lett. 36(6), 822–824 (2011). [CrossRef] [PubMed]
18. F. P. Mezzapesa, L. Columbo, M. Brambilla, M. Dabbicco, A. Ancona, T. Sibillano, F. De Lucia, P. M. Lugarà, and G. Scamarcio, “Simultaneous measurement of multiple target displacements by self-mixing interferometry in a single laser diode,” Opt. Express 19(17), 16160–16173 (2011). [CrossRef] [PubMed]
19. F. De Lucia, M. Putignano, S. Ottonelli, M. di Vietro, M. Dabbicco, and G. Scamarcio, “Laser-self-mixing interferometry in the Gaussian beam approximation: experiments and theory,” Opt. Express 18(10), 10323–10333 (2010). [CrossRef] [PubMed]
20. S. Ottonelli, M. Dabbicco, F. De Lucia, and G. Scamarcio, “Simultaneous measurement of linear and transverse displacements by laser self-mixing,” Appl. Opt. 48(9), 1784–1789 (2009). [CrossRef] [PubMed]
22. A. E. Wynne and B. C. Stuart, “Rate dependence of short-pulse laser ablation of metals in air and vacuum,” Appl. Phys., A Mater. Sci. Process. 76(3), 373–378 (2003). [CrossRef]
23. A. Ruf, P. Berger, F. Dausinger, and H. Hugel, “Analytical investigations on geometrical influences on laser drilling,” J. Phys. D Appl. Phys. 34(18), 2918–2925 (2001). [CrossRef]