CA2296924A1 - Burst-ultrafast laser machining method - Google Patents

Abstract

A method of laser processing of materials. The combination of ultrafast laser pulses and high-repetition rate (>1MHz) bursts (or continuous operation) defines a new regime for material processing. The high rate controls thermal processes evolving between each ultrafast laser pulse that 'prepares' the sample surface or bulk for optimal interaction with subsequent ultrafast laser pulses.
The addition of this laser-controlled thermal component overcomes several current limitations of ultrafast laser processing at lower repetition rates (<
1MHz), providing means to fully harness the many attributes by ultrafast lasers for material processing applications.

Description

BURST-ULTRAFAST LASER MACHINING METHOD
FIELD OF THE INVENTION
The present invention relates to methods of laser processing of materials, and more particularly the present invention relates to laser processing of materials using ultrafast laser pulses.
BACKGROUND OF THE INVENTION
Many efforts in the current generation of laser processing of materials can o be described as investigating new modalities in which the laser fluence may be delivered to a workpiece, specifically the ways in which the pulse duration, wavelength or pulse-shape give significant new control over the laser-material interaction.
Various studies have shown that laser material processing in the ultrashort-pulse regime offers numerous advantages compared with longer pulses, see for example J. X. Zhao, B. Huttner, and A. Menschig, SPIE Proc Vol. 3618, (1999);
C.
Momma et al., Appl. Surf. Sci., 109/110,15 (1997); C. Momma et al., Optics Comm., 129, 134 (1996); D. von der Linde, K.Sokolowski-Tinten, and J. Bialkowski, Appl.
Surf. Sci. 1091110, 1 (1997); and X.Liu, D.Du, and G. Mourou, IEEE J. of Quantum 2o Electron. 33, 1706 (1997). High intensity is available to micromachine and process surfaces cleanly by aggressively driving multi-photon, tunnel ionization, and electron-avalanche processes, see J. Ihlemann, Appl Surf. Sci. 54 (1992) 193;
D. Du, X. Liu, G. Korn, J. Squier, and G. Mourou, Appl. Phys. Lett. 64 (1994) 3071;
W. Kautek and J. Kruger, SPIEProc. 2207 (1994) 600; P. P. Pronko, S.K. Dutta, J.
Squier, J.V. Rudd, D. Du, G. Mourou, Optics Comm. 114 (1995) 106; B.C. Stuart, M.D. Feit, S. Herman, A. M. Rubenchick, B.W. Shore, M.D Perry, J. Opt. Soc. Am 813 (1996) 459; and C.B. Schaffer, A. Brodeur, N. Nishimura, and E. Mazur, SPIE
3616 (1999) 143.
Absorbed fluence must also be controlled, and shorter pulses in the picosecond and femtosecond regime therefore become necessary. Ultrafast laser interactions have well-defined 'damage' thresholds offering improved precision in processing applications, see X.Liu, D.Du, and G. Mourou, IEEE J. of Quantum Electron. 33, 1706 (1997) and D. Du, X. Liu, G. Korn, J. Squier, and G.
Mourou, Appl. Phys. Lett. 64 (1994) 3071. Much recent literature has been devoted to ultrafast laser damage and processing of transparent or wide-bandgap materials, see J. Ihlemann, Appl Surf. Sci. 54 (1992) 193, D. Du, X. Liu, G. Korn, J.
Squier, and G. Mourou, Appl. Phys. Lett. 64 (1994) 3071. Nonlinear absorption mechanisms are key to coupling laser energy into such non-absorbing media.
o Thermal impact during picosecond and femtosecond laser interactions is highly limited, confining laser energy dissipation to small optical penetration depth with minimal collateral damage. This precisely confined laser 'heating' minimizes the energy loss into the underlying bulk material, providing for an efficient and controllable ablation process, see S. Preuss, A. Demchuk, and M. Stuke, Appl.
~5 Phys. A, 61, 33 (1995); and T. Gotz and M. Stuke, Appl. Phys. A, 64, 539 (1997).
With tight focusing, small interaction volumes on the scale of ~1 ~rm3 is possible which enhances evaporative cooling since the fractional volume change on expansion becomes very large. The short duration further ensures that, all of the laser energy arrives at the surface before the development of an ablation plume 2o and/or plasma; such efficient energy coupling is not available with longer duration (>10's ps) laser pulses because of plasma reflection and plume heating. Such ultrafast-processing features are highly attractive for the precise microprocessing of good heat conductors such as metals; at the same time, nonlinear absorption of these intense ultrafast pulses also reduces the ablation threshold for wide-bandgap 25 or "transparent" optical materials such as silica glasses.
Ultrafast lasers also offer the means to internally process transparent glass.
Microexplosions provide opportunities for 3-D optical storage (C.B. Schaffer, A.
Brodeur, N. Nishimura, and E. Mazur, SPIE 3616 (1999) 143)while refractive index structures such as volume gratings and waveguides (K.M. Davis, K. Miura, N.

Sugimoto, and K. Hirao, Opt. Lett. 21 (1996) 1729) have been formed in which the local index of refraction can be permanently altered.
While ultrafast lasers offer exciting prospects for processing materials, at present undesirable effects exist and processing windows are poorly defined.
Effects requiring more control include in laser processing of materials includes incubation (defect generation) effects that change etching rates, self focusing and clouding effects, 'gentle' and 'strong' ablation phases developing with increasing number of pulses, pre-pulse or pedestal effects, poor morphology: periodic surface structures, melt, debris, surface swelling, shock-induced microcracking, slow o processing rates and saturation of hole depth in via/hole formation.
It is advantageous to provide a method of laser processing of materials that addresses the aforementioned difficulties present in present processing methods.
SUMMARY OF THE INVENTION
~5 The present invention provides a method of processing materials based on high-repetition rate (continuous or burst) application of ultrafast laser pulses to materials. More particularly the present invention provides a novel method of controlling the delivery of laser fluence to a material during laser processing that reduces unwanted damage in the material.
2o In one aspect of the invention there is provided a method of laser material processing, comprising providing a material to be processed and applying laser pulses to a target zone on the material, the laser pulses having a time separation between individual laser pulses sufficiently long to permit hydrodynamic expansion of a plume and/or plasma so that a next subsequent laser pulse is not substantially 25 reflected, scattered and/or absorbed by the plume and/or plasma, and the laser pulses having a time separation between laser pulses sufficiently short so that a thermal component in the target zone presents heated material to successive laser pulses.
The laser pulses may be applied at rates above 1 MHz wherein thermal transport does not completely remove heat deposited and/or transported in or near the processing volume by each laser pulse. A sheath of heated material is therefore retained and presented to each subsequent laser pulse.
This thermal component offers a new modality for controlling ultrafast-laser processing. By adjusting the pulse-to-pulse separation (inverse of repetition rate), the temperature rise, and the size of the heated sheath is controlled. This material heating offers several advantages and opportunities that are not available for material processing at lower repetition rate. An increased temperature dramatically alters the materials properties in a manner that can positively affect the ultrafast o interaction, and control subsequent events such as shock development, defect formation, annealing, surface morphology, debris formation, plume evolution, material removal rates, and geometry of excisions. The combination of high-repetition rate with ultrafast laser pulses provides added control and new avenues in material processing.
BRIEF DESCRIPTION OF THE DRAWINGS
The following is a description, by way of example only, of the method of laser processing of materials in accordance with the present invention, reference being had to the accompanying drawings, in which:
2o Figure 1 a shows an atomic force microscopy (AFM) image of a micro-hole in fused silica, drilled by a single 1.2-ps laser pulse with a peak fluence of 9.1 J/cm2;
Figure 1 b shows a depth profile corresponding to the hole shown in Figure 1 a;
Figure 1c shows an atomic force microscopy (AFM) image of a micro-hole in fused silica, drilled by a single 1.2-ps laser pulse with a peak fluence of 38 J/cm2 (bottom);
Figure 1d shows a depth profile corresponding to the hole shown in Figure 1 c;

Figure 2 shows a plot of excised hole depth as a function of accumulated laser fluence (i.e., number of laser pulses) for single-pulse fluences of 9.6 and 31 J/cm2;
Figure 3 shows a plot of etching depth per pulse in fused silica as a logarithmic function of laser fluence;
Figure 4a shows a series of optical microscope photographs of fused silica ablated by 1.2-ps Nd:glass pulses at 140 J/cm2, from left to right, holes were drilled by one, two, three, four, and five pulses;
Figure 4b shows two SEM photographs at two different magnifications of o fused silica showing features of the shock-induced microcracks, the holes were ablated with four pulses at 93-J/cm2 fluence 00.06 Hz);
Figure 5 shows an SEM angle view of hole excised in BK7 glass by a mode-locked pulse train consisting of 250 single 1.2-ps laser pulses with a pulse-to-pulse separation of 7.5-ns;
~5 Figure 6 shows etch depth in BK7 glass plotted as a logarithmic function of the total burst laser energy;
Figure 7 shows the number of shots (pulse-trains) to drill through aluminum foils of 25 mm and 100 mm thickness as a function of the pulse-train fluence;
2o Figure 8 shows etch rates per pulse-train burst for various foil thicknesses of aluminum;
Figure 9a,shows an SEM photograph of a hole drilled through 200-mm thick aluminum foil (laser-irradiated surface) with one pulse-train burst at 3.16 kJ/cm2 fluence;
25 Figure 9b shows an SEM photograph of a hole drilled through 200-mm thick aluminum foil (rear surface) with one pulse-train burst at 3.16 kJ/cmz fluence;
Figure 10a shows an SEM photograph of laser irradiated surfaces comparing two holes drilled through a 50-mm thick aluminum foil: left hole:

fluence 480 J/cm2, three bursts followed by one 'cleaning' shot; . right hole:
fluence 5.36 kJ/cm2, one shot plus one 'cleaning' shot;
Figure 10b are SEM photographs of the holes of Figure 10a but taken from the rear surface of the holes with the left (right) hole corresponding to the right (left) hole in Figure 10a; and Figure 11 shows a plot of observed etch-depths into a 200 pm foil, compared to calculated vaporization depth.
DETAILED DESCRIPTION OF THE INVENTION
o The method of ultrafast laser machining of materials will be exemplified with two illustrative non-limiting examples. Detailed examples are provided here for finro classes of materials including brittle transparent glass and ductile metal aluminum. It will be understood that the principles demonstrated herein are extensible to a large range of material classes for broad application in ultrafast-~5 laser material processing.
Laser Systems For applications directed at material processing, ultrafast-laser systems presently available typically combine a mode-locked oscillator with an amplifier that raises the single-pulse energy to levels suitable for material modification.
2o While such oscillators provide high repetition rates (~100MHz continuous or in bursts), practical considerations in the amplifier power have precluded the amplification of each oscillator pulse. Only a small number of the oscillator pulses are amplified in ultrafast laser systems currently employed in material processing research and development (typically at rates much less than 1 MHz).
25 For these considerations, thermal diffusion between laser pulses at such low rates provides sufficient time to transport away most or all of any thermal energy deposited by the laser into the volume immediately surrounding the processing region. This heat transport cools the sample surtace to that of the underlying bulk material before the arrival of the next ultrafast laser pulse.

The present invention makes use of lasers in the high-repetition-rate ultrafast-laser processing of materials. The results described here are not particular to the laser system described below, but share a common physical process/interaction with all ultrafast laser system operating at high repetition rate (>1 MHz).
A feedback-controlled Nd:glass oscillator (~ = 1054 nm) operating at 1 Hz repetition rate provided a near flat-topped train of 430 mode-locked pulses (quasi cw) with pulse-to-pulse separation of 7.5 ns, see R.S. Marjoribanks, F.W.
Budnik, L. Zhao, G. Kulcsar, M. Stanier, & J. Mihaychuk, Optics Lett. 18, 361 o (1993). A single high-contrast pulse of 1.2-ps duration was selected by an external Pockets cell and amplified 13-fold in a four-pass geometry at 0.06 Hz (limited by our flashlamp). The ~3-NJ pulses were focussed by interchangeable aspherical lenses (AR-coated BK7; f = 15.4, 11.0, 4.5 and 3.1 mm) to near diffraction-limited spot sizes of 3.2, 2.0, 1.0, and 0.8 pm (1/e2) diameter, ~5 respectively. The test samples (UV-grade fused silica Corning 7940; BK7 glass, aluminum foil) were mounted on a precision x-y-z stage. Focusing was monitored by image-relaying the retro-reflected beam from the focal spot, with magnification, onto a CCD camera. On-target fluence was varied over the range 2 to 170 J/cm2 by adjusting the amplifier gain, using neutral density filters, and 2o employing different focal-length lenses. Excisions were made using between and 60 pulses, of various fluence values. All samples were irradiated in air, and transverse nitrogen gas flow was at times used to reduce the accumulation of ablation debris. Self-focusing effects in air were not seen, at this pulse duration, peak power, and with the short focal-length lenses used. Laser focusing 2s conditions also did not produce bulk discoloration or damage effects in glass regions beneath the excised holes (as evidenced by optical microscopy).
The burst mode was provided by a waveplate which passed the full oscillator train of several hundred pulses for high-repetition-rate machining at 133 MHz. The ~3-ps long pulse train was amplified and focussed onto glass or metal surfaces as described above, accumulating a total fluence of ~40 kJ/cm2 in a diameter of ~2 pm. The burst duration was varied from 250 to 430. The resulting ultra-high repetition rate (133 MHz) pulsetrain had a nearly flat waveform, with a risetime of about 100 ns and a falltime of about 500 ns. With s the four-pass amplifier optional, pulse-train energies of 0.05 to 2 mJ were available for all studies.
By employing various electrooptic devices, and alternate cavity and/or amplifier designs, tuning of the single-pulse duration, pulse-to-pulse separation, duration of burst (to cw), and the temporal profile of the burst envelope becomes o available to optimize and control the laser-material interaction and subsequent processes in the sample material.
Prior Art Method Of Low Repetition Ultrafast Processing Of Glass In this section,. results of 1.2-ps laser ablation of fused silica and BK7 at ~5 repetition rates of 1 Hz or less are described as a reference to compare with the attributes of burst machining forming the present invention described in the next section. Fused silica and BK7 are highly transparent at the 1.05-pm laser wavelength and yield similar micromachining results. Surface morphology of microholes formed by single laser-pulses are shown in the AFM photographs in 2o Figures 1 a and 1 c. Fluences of 9.1 and 38 J/cm2 each produced moderately smooth holes of --2.0-pm diameter (FWHM). Surface-profile traces, shown in Figures 1 b (corresponding to Figure 1 a) and Figure 1 d (corresponding to Figure 1 c) , reveal hole depths of 100 and 360 nm, respectively.
A small ring structure is observable in the higher-fluence hole, a feature also 25 reported by D. Ashkenasi, H. Varel, A. Rosenfeld, F. Noack, and E.E.B.
Campbell, Nucl. Instr. & Meth. in Phys. Res. 8 122, 359 (1997) for 3.2-ps ablation of fused silica. The excised surface contour was found to crudely follow the laser beam profile, with small-scale surface roughness of t10% (rms) of the hole depth.
This t10% surface roughness was a general observation for the 'gentle' ablation phase, noted here for fluences, F, less than ~44 J/cm2. Even when several pulses were applied to the same area, surface roughness typically increased in absolute terms, but remained limited to ~10% of the final hole depth.
Figure 2 shows the progress of hole depth with the number of laser pulses, N, for fluence values of 9.6 and 38 J/cm2. (Accumulated fluence was used for the abscissa to better account for the ~10% variations of the laser energy). For 9.6 J/cm2 (38 J/cm2), the depth increases linearly with N (accumulated fluence) to an apparent peak value of 2.7 Nm (2.2 pm) after 14 (6) pulses. The nominal plateau apparent for higher pulse-number is simply an artifact of the AFM tip, which cannot o probe larger aspect-ratio holes: those deeper than their --2-Nm diameter. It is anticipated that the hole depth will in fact rise linearly with number of laser pulses through and beyond this plateau region until incubation processes raise the density of defects or color centers to a critical value. Beyond this critical value, the 'gentle' ablation process is expected to give way to 'strong' ablation, a distinct regime ~5 wherein etch rates (depth per pulse) can be increased more than 10 fold, see A.C.
Tam, J.L. Brand, D.C. Cheng, and W. Zapka, Appl. Phys. Lett. 55, 2045 (1989).
For fused silica, there have been reported etch rates of 550 nm per pulse when 100's of pulses at 1.3-ps duration were applied at 12 J/cm2 fluence, see H.
Varel, D. Ashkenasi, A. Rosenfeld, R. Herrmann, F. Noack, E.E.B. Campbell, Appl.
Phys.
2o A 62, 293 (1996). Their rate is triple this 180 nm/pulse rate for the same fluence, but with N < 10. For reasons given below, the transition to strong ablation with increasing N was not studied here.
The etch-depth data in Figure 2 show that material removal was initiated with the first laser pulse (N = 1 ), for both 9.6 and 38 J/cm2 fluences. Incubation effects 2s developing at fluences before the onset of ablation were not studied in the present work although such effects are already anticipated below our single-pulse ablation threshold of ~5.5 J/cm2. Kautek et al. reported the need for 7 incubation pulses before low-fluence (1 J/cm2) ablation of barium-aluminum borosilicate glass could proceed with 50-fs laser pulses, see W. Kautek, J. Kruger, M. Lenzner, S.
Sartania, C. Spielmann, and F. Krausz, Appl. Phys. Lett. 69, 3146 (1996). Such incubation processes are undesirable for practical applications, and removing control over etching rates.
Single-pulse etch rates were collected from the slopes of data in graphs like s Figure 2 and plotted in Figure 3 as a function of single-pulse fluence. Two regimes, gentle and strong ablation, are identified. Representation of the data (solid lines) by (1/ae~) log (FIFt,~ provide values for threshold fluence and the effective absorption coefficient in each regime. The etch-rate data follow a logarithmic fluence-dependence from an extrapolated ablation threshold of 5.5 J/cm2 to ~44 o J/cm2, the onset of strong ablation. This fluence window (5.5 to 45 J/cm2) defines the gentle-ablation processing window for controllable etching of smooth features in fused silica. Thin layers, 100-nm deep or less, could be accurately excised with appropriate choice of fluence. The logarithmic fluence dependence, normally associated with single-photon absorption mechanisms. is sur~risinn nPrA
~5 considering the nonlinear mechanisms that are understood to drive absorption in this transparent material. Kautek et al. have also reported a logarithmic fluence dependence for 20-fs to 3-ps ablation of barium aluminum borosilicate glass (W.
Kautek, J. Kruger, M. Lenzner, S. Sartania, C. Spielmann, and F. Krausz, Appl.
Phys. Lett. 69, 3146 (1996)). For 1.2-ps ablation of fused silica, the slope of the 2o solid curve in Figure 3 (for F < 44 J/cm2) provides an effective penetration depth of 1/ae,~= 235 nm, a value commensurate with the 100-nm layer-by-layer resolution cited above. The 5.5 J/cm2 threshold fluence is in accord with the damage threshold of 5t1 J/cm2 reported by Varel et al.for 1.0-ps ablation of fused silica, see H. Varel, D. Ashkenasi, A. Rosenfeld, R. Herrmann, F. Noack, E.E.B. Campbell, Appl.
Phys.
25 A 62, 293 (1996). This group also report in a later paper, etch rates of nm/pulse with 3.2-ps pulses at 10 J/cm2 fluence, see D. Ashkenasi, H. Varel, A.
Rosenfeld, F. Noack, and E.E.B. Campbell, Nucl. Instr. & Meth. in Phys. Res. B
122, 359 (1997). This etch rate is only slightly larger than our 180 nm/pulse value from Figure 3 for 1.2-ps pulses. Note again that the rates in Figure 3 are only valid where the number of laser pulses is small. The onset of a strong ablation phase after 10's or 100's of laser pulses explains the 2 or 3-fold faster etch rates reported for ablation of deep channels in fused silica in D. Ashkenasi, H. Varel, A.
Rosenfeld, M. Whamer and E.E.B. Campbell, Appl. Phys. A 65, 367 (1997).
A higher fluence, F > 44 J/cm2 in Figure 3, etch rates abruptly rise to values 2- or 3-fold faster than by simple extrapolation of the gentle ablation data.
This enhanced rate is related to the incubation phenomenon described previously where now a single pulse provides sufficient fluence to fully incubate the underlying glass material. The effective penetration depth rises to 9/ae,~= 780 nm, supporting rapid o etch rates of up to 2-Nm per pulse at a fluence of 150 J/cm2. Such rapid etch rates are attractive for many applications, however, this strong-ablation regime provides less control over etch depth as evidenced by the wider scatter of data points in Figure 3. A further disadvantage is the development of microcracks following 2 or 3 ablation pulses at high fluence as discussed below.
~5 While these low-repetition-rate ultrafast laserobservations appear promising for controllable etching of optical materials, detrimental effects are noted.
Most significant is the development of shock-induced microcracks, and shearing and flaking of surrounding surfaces following a small number of moderate-intensity pulses. Figures 4a and 4b shows the rapid development of shock-induced 2o microcracks forming around the perimeter of laser-ablated holes. Figure 4a shows a series of optical microscope photographs of fused silica ablated by 1.2-ps Nd:glass pulses at 140 J/cm2, from left to right, holes were drilled by one, two, three, four, and five pulses. Figure 4b shows two SEM photographs at two different magnifications of fused silica showing features of the shock-induced microcracks, 25 the holes were ablated with four pulses at 93-J/cm2 fluence 00.06 Hz).
Shock-induced microcracks developed quickly, by the third pulse, for this large fluence.
At the 140 J/cm2 fluence, microcracks and surface swelling (noted by AFM
and SEM) developed very quickly by the third laser pulse thus posing a significant limitation to precise shaping of smooth optical surfaces. At lower fluence, microcracking developed more slowly. Over the 5.5 to 170 J/cm2 fluence-window studied here, these undesirable surface features appeared consistently after an onset number of laser pulses, N~, that approximately followed N~ = 1.7 + 80/F (F in J/cm2).
Because these N~ values are small, peaking at -25 near the threshold fluence for the gentle ablation region, there was no practical reason for extending studies to integrate large numbers of pulses (N > 60). Therefore, the transition from gentle ablation to strong ablation with increasing N was not observed here, preceded by the early development of microcracks, the main limitation to smooth surface-o structuring of fused silica. D. Ashkenasi, H. Varel, A. Rosenfeld, M. Whamer and E.E.B. Campbell, Appl. Phys. A 65, 367 (1997) also reported the formation of microcracks around deep (~1 mm) channels etched in fused silica by hundreds of laser pulses of 100-fs to 30-ps duration. Their study showed a favorable trend of reduced microcracking with decreasing pulse duration.
~5 Combination of N~ in Equation 1 with the per-pulse-etch rates in Figure 2 provides a coarse guide to the maximum ablation depth one can attain without deleterious microcracking or surface swelling phenomena. Structures up to ~1-Nm deep with 10% rms surface roughness are shown here to be possible, establishing a practical but very restricted processing window for ultrafast-laser micromachining 20 of fused silica and related transparent materials.

Burst Ultrafast Processing Of Glass The full flat-topped oscillator pulse train comprising 250 and 400 pulses of 25 1.2-ps duration at 5 to 150 J/cm2 fluence each, was applied to fused silica and BK7 surfaces. The optical SEM photo in Figure 5 shows a high aspect-ratio via formed as a result of this single pulse train. A smooth symmetric hole of ~10-Nm diameter was excised to 15-pm depth (determined with optical microscopy). The ~15-mm deep hole has smooth walls and shows no evidence of fractures, cracks, or collateral damage. Only a small mount of ejected melt has solidified on the entrance hole perimeter. The entrance hole diameter of ~14 mm exceeds the focused laser beam diameter of ~1.8 microns. The burst energy was 1.48 mJ and the total (integrated) fluence was ~49 kJ/cm2.
In comparison, the low repetition rate (1-Hz) result described in the previous section showed that microcracks formed after only 3 pulses for the same single-pulse fluence. Cumulative heating effects associated with the 133-MHz pulse-repetition rate in the pulse train are proposed to improve the ductility of the surrounding glass, thereby mitigating the shock-induced microcracking in regions o immediately surrounding the hole perimeter. Such heating effects also support an annealing effect. Heat incompletely dissipated on a nanosecond time-scale may account, in part, for the expansion of hole diameters to 8-10 Nm which is ~5?
larger than the diameter of the focussed laser beam. For fluences well above threshold, ablation also extends appreciable into the weaker edges of the Gaussian beam.
~5 In the operation of the feedback-controlled mode-locked oscillator, the first dozen pulses at the leading edges of the train are also somewhat longer-duration (~10 ps); possibly this may also have an effect.
Etch depths excised in BK7 are plotted in Figure 6 as a function of total energy in the 250-pulse envelope. A single train of 250 pulses of 1.2-ps 2o duration each with a pulse-to-pulse separation of 7.5 ns was applied.
Single bursts were applied to one surface of a 90° prism and etched depths were measure from the adjacent with an optical microscope. The data are coarsely represented by a logarithmic energy dependence with an extrapolated ablation threshold of ~15 NJ energy or 500 J/cm2 fluence and an effective absorption 25 coefficient of 2900 cm-' (from the inverse slope). The latter value yields an effective optical penetration depth of 3.4 Nm for the full train of pulses, a value 4-fold (15-fold) larger than the corresponding value for strong (gentle) ablation regime in Figure 3 for fused silica.
Multiple pulse trains were applied to fused silica at.1-Hz repetition rate, resulting in a slight increase (~20%) in hole diameter and a moderate advance in etch depth to several 10's of microns. Deeper holes are anticipated with modification of the laser parameters and the focusing geometry. Hole depth saturated quickly for 100-J/cm2 single-pulse fluence. Saturation of hole depth when drilling deep channels in fused silica with femtosecond and picosecond laser pulses at 10-1000 Hz repetition rate has been previously reported, see D. Ashkenasi, H. Varel, A. Rosenfeld, M. Whamer and E.E.B. Campbell, Appl.
Phys. A 65, 367 (1997). However, unlike these lower repetition rate results, the application of many bursts in the method disclosed herein has not led to the o formation of microcracks, fractures, or swelling for any samples in the present work, a very surprising and important advantage for shaping smooth surface structures, especially high aspect ration holes and blind vias.
High repetition-rate multi-pulse ablation disclosed herein is clearly a promising new option for controlling the micromachining quality of brittle ~5 materials. The 7.5-ns pulse-to-pulse separation used herein is sufficiently short to reduce the material cooling between laser pulses, thereby permitting the presentation of a heated and more ductile glass to succeeding laser pulses in the small processing volume. During the 7.5-ns interval between pulses, the thermal diffusion scale length, (4D~)'~2, is 0.17 pm in glass, a value surprisingly 2o comparable with the effective optical penetration depth of 0.25 Nm in the single-pulse gentle ablation regime (Figure 2). Since the plume will carry not all absorbed laser energy away, retention of this dissipated energy within a scale length comparable with the laser penetration depth ensures that subsequent laser pulses interact with a thermally modified glass while minimizing the heat-25 affected zone. An important additional consideration of the pulse-to-pulse separation is to provide sufficient time for hydrodynamic expansion and dilution of the laser-produced plume and plasma, reducing or eliminating obscuration of subsequent laser pulses. This is an important benefit that retains the advantages of ultrafast-laser material processing (i.e. laser dissipation in the bulk material) while also offering control of the heat retained in the nearby laser-interaction volume of the material. The pulse-to-pulse separation becomes an important new optimization parameter, controlling the amount of laser heat retained in the sample (higher temperature when reduced) and the amount of laser energy lost to an incompletely dissipated plume (less loss when increased). This control and these general advantages are available to brittle materials in general, and include but not limited to ceramics, tooth enamel, bone, and composite materials for a wide range of applications.
o EXAMPLE 2 Burst Ultrafast Processing Of Aluminum High-repetition rate burst machining was applied to aluminum (Goodfellow, 99%) foils of thicknesses of 12.5, 25, 50, 100 and 200 Nm.
Samples were mounted free-standing to preclude effects of heat conduction into ~5 any substrate. A photodiode was placed directly behind the foil to signal the laser burnthrough of the foil. On-target laser energy was controlled by neutral density filters and amplifier gain. Ablated surfaces were examined by scanning electron microscopy (SEM), atomic-force microscopy (AFM), and optical microscopy.
2o The number of laser pulses necessary to drill through 25-Nm and 100-Nm foils are plotted in Figure 7, as a function of pulse-train fluence. The fluence values are divided by 250 to obtain the single 1.2-ps pulse fluence. A large fluence range of 80 to 9000 J/cm2 was examined. Qualitatively, the ablation behaves as expected: the minimum number of bursts required to drill through a 25 foil increases with increasing foil thickness and decreases with increasing fluence. For each thickness there was a threshold fluence below which the target could not be pierced by even a hundred shots, even though this fluence was itself well above the damage threshold at the surface. This is understood to be related to a reduction of etch rate with depth. In the measurements disclosed herein this piercing threshold increases with foil thickness, from 120 J/cm2 for 25 Nm foils to 600 J/cm2 for 100-pm foils. This ~4-fold difference in threshold is attributed in part, perhaps, to distributed absorption along the length of increasingly deep channels, and to increasing heat losses into the bulk sample on the 2 Ns time scale of the laser burst. Likely the effect. of imperfect waveguiding in the channels also reduces laser intensity at the hole-bottom.
Such losses raise the material-removal threshold fluence as increasingly deeper channels are bored out by the laser. Beyond a maximum fluence of --200 and 6900 J/cm2 for 25 and 100-Nm foils, single bursts will cut through the foil.
Except o for single-shot piercing data, pulse-to-pulse energy fluctuations of ~30%
lead to a scatter of data points especially near the through-hole fluence threshold.
Burnthrough etch-rate data are presented in Figure 8 for each of the foil thicknesses tested. Etch depths were interpreted from graphs such as Figure 7, identifying the minimum fluence necessary to reproducibly punch through the foil for a given number of pulses, then plotting against that fluence the average etch rate per pulse (from the foil thickness and number of pulses to pierce). All foils except the 200-Nm thickness could be consistently drilled through with a single burst. The etch-depth data are strongly dependent on foil thickness.
Figure 9 shows SEM photographs of the front and the back surfaces of 2o the 200-Nm foil that was drilled with one burst at 3.16 kJ/cmz. The hole perimeter is relatively clean at both surfaces with only a thin (~3-Nm) wide rim of melt splattered around the entrance hole. No optimization effort was made to minimize this splatter. The entrance hole diameter of 30 Nm reduces to 7.5 Nm at the back side, yielding a 7:1 aspect-ratio hole with tapered sides at ~3° on either side of target normal. The aspect ratio could be adjusted with changes to the laser fluence and focusing conditions.
Figure 10 illustrates the influence of the laser fluence on the hole diameter. The left hole was drilled with three pulse-train bursts, each at a fluence of 480 J/cm2, followed by a fourth shot of the same fluence to trim the hole of any melt/flow irregularities inside. The entrance hole diameter is ~6 Nm which corresponds closely to the 5.6 Nm laser-beam diameter. A comparatively large amount of re-solidified material is also seen to surround the hole perimeter. The laser fluence is only ~50% above the minimum fluence required to produce a through-hole for this case (240 J/cm2) and melt processes appear to reduce the hole quality. Note that this laser fluence is approximately an order of magnitude above the surface damage threshold. A SEM photo of the rear side of this through-hole is shown on the right in Figure 10b.
The right hole in Figure 10a was drilled with a single burst at a fluence of o 5.36 kJ/cm2, followed by an additional shot to clean out the drilled hole of debris. The 11-fold higher fluence produced a larger hole diameter of ~30 Nm, about 5 times the laser focal spot size (FWHM) and producing a hole 25 times the area. The rear side of the hole, shown on the left in Figure 10b, is also much larger (~20 Nm) in diameter than the lower-fluence example. Assuming linear ~5 absorption, the increase in hole diameter is commensurate with the increase in laser power: for a Gaussian profile, the intensity of the laser has near the same value at the hole-edge in each case. This argues for a local threshold effect, such as the specific energy in the target material passing that amount needed for melting. Figures 9 and 10 demonstrate control over hole diameter and 2o aspect ratio with laser fluence in burst laser machining of metal foils.
The absence of substantial melt debris, especially for fluences 10x to 100x above the surface-damage threshold, demonstrates that long-pulse physics dominated by melt-phase material ejection is not taking place here. However, during such a microsecond pulse-train, heat will have diffused into the material 25 surrounding the laser spot in a manner similar to that described above for the glass studies. A simple consideration of the diffusion length, (4D~)'~2 provides a heat scale length of a fraction of a micron for the 7.5-ns interval between picosecond pulses in the train, and 28 Nm over the whole 2-Ns pulse-burst. While this rule-of-thumb can be misleading in assessing the hole diameter, it demonstrates a compact scale length over which heated material is presented to each ultrafast pulse within the train. Below 1 MHz repetition rates, much of the heat retained by the sample will have diffused into the underlying bulk material, lowering the sample surface temperature to that of the bulk. At 133 MHz, the ultrafast interaction takes place within a heated zone of the material. In this way, deeper channels can be excised because of a reduced ablation threshold, improving the energy efficiency of the material removal.
Through holes can excised in thick foils with single bursts, greatly improving machining time over that provided by traditional sub-MHz lasers. The thermal o component also affords control over the diameter and aspect ratio of the hole.
These benefits are in addition to those normally associated with ultrafast laser processing.
The thermal physics of pulse-train burst interaction is therefore intermediate between that of single long-duration and ultrafast pulses. It ~5 appears that there are some advantages of heating or annealing surrounding material without the gross melting characteristic of longer-duration pulses.
Likely this is because ultrafast laser pulses have the advantage of evaporative cooling, over a hydrodynamic timescale of the expanding plume, as the locally heated material vaporizes and expands away from the solid, decoupling from it 2o thermally. In this case, much of the heat impulse of an ultrafast laser pulse is carried away with the plume/plasma, producing etching more similar to material sublimation than is possible for quasi-cw machining. As a result, the characteristic heating time is limited to the timescale of hydrodynamic expansion (and evaporative cooling). For this reason, it appears that the etched hole-size 25 is fairly closely linked to the local specific energy deposition by the laser, as it compares to the specific energy of vaporization, and less by lateral thermal transport.
This is supported by the results of Figure 11. Figure 11 shows a plot of observed etch-depths into a 200 pm foil, compared to calculated vaporization depth, the solid line marks the deepest holes thermodynamically possible, if 30%
of the incident laser energy were invested in vaporizing the material immediately underlying the focal spot, i.e., without considering any lateral-transport effects.
This overestimates the depth: using the observed (larger) hole-diameters, instead of the laser spot-size, would bring this available-energy vaporization depth closer to quantitative agreement with the data, as would smaller absorption fractions.
Vaporization of aluminum will take place with an energy dissipation ~36 kJ/cm3 (includes thermal capacity, heat of fusion, and heat of vaporization).
o Setting aside the laser absorption efficiency, this value is used to plot a curve of maximum depth attainable by evaporative ablation, D, as a function of fluence:
D = F / (3.6 kJ/cm3). Since all the measured data fall below this curve, we see that mode-locked laser bursts nominally can provide more than adequate fluences to vaporize aluminum to the observed depths. Corrections for ~5 absorption-factor, and for ratio of observed hole-size to nominal laser spot-size, bring closer quantitative agreement. Therefore, the 1.2-ps laser-matter interactions appear to drive a vaporization-phase ablation process with the commensurate advantages of little melt-debris and improved resolution that is not available with nanosecond or microsecond interactions. A vaporization 2o process was similarly inferred by Zhu et al. for single-pulse femtosecond ablation of aluminum, see X. Zhu, A. Naumov, S. Nikumb, P. Corkum, Drilling sub-10 micron holes in metal foils with femtosecond laser pulses, to appear Appl. Physics A.The 133-MHz rates used here therefore retains the advantages of ultrafast laser microprocessing.
25 Summarizing for aluminum, it has been observed that drilling of smooth and relatively clean, high aspect-ratio through-holes in foil thicknesses up to 200 pm with single microsecond bursts provided a faster process than possible with current kHz repetition-rate systems. As a hybrid way of delivering laser-fluence to target, these mode-locked bursts exploit an excellent combination of quasi-cw heating effects to support rapid etching rates and ultrafast-laser interactions for clean ejection of material. Plume-absorption effects are also mitigated to the degree that the, 7.5-ns pulse-to-pulse separation supports hydrodynamic expansion of ablation vapor/plasma from the surface. In this way, large aspect-ratio holes could be formed in thick metal foils with a single burst.
Ultrafast laser interactions and thermal diffusion are similar for metals and semiconductors so that the inventors contemplate the general advantages of the present burst-ultrafast processing of aluminum are therefore extensible to this broad and general class of materials. The results of laser machining of glass o entails the other extreme of material properties (brittle and high melt/vapour temperatures), demonstrating that burst-ultrafast lasers offer a wide spectrum of applications and advantages in laser material processing of materials having widely diverse properties.
It will be understood that the method of the present invention may be used ~5 for processing in the bulk of the material and is not restricted to processing the surface of the material. In this case the process involves applying laser pulses to a target zone within the body of the material, the laser pulses having a time separation between individual laser pulses sufficiently long to permit acoustic and thermal shock to the material to spread and/or dissipate so that a next 2o subsequent laser pulse is not substantially reflected, scattered and/or absorbed by the temporarily altered material properties. The laser pulses have a time separation between pulses sufficiently short so that a thermal component in the target zone presents heated material to successive ultrafast laser pulses in the burst, control of which residual or accumulated heating serves the purpose of 25 preventing or mitigating against the deleterious effects of material stresses in the material due to,acoustic or thermal shock, while optimizing the useful range of such effects.
The method disclosed herein defines a new way of controlling the delivery of laser fluence to optimize performance by utilizing the attributes of ultrafast laser interactions with advantages of long pulse heating. The heat-induced stresses caused by thermal cycling of known up-to-kHz repetition rate machining include 'bound' stresses, normally caused by laser-heating. If the material cools down between pulses, stresses are locked into glasses and ceramics. Then the third pulse may cause brittle fracture. The advantages of the present invention include specifically that the high repetition rate avoids thermal cycling (calculate by thermal diffusion times, roughly); also that the high repetition rate anneals thermal stresses pre-existing or accumulating in the material. Specific to the present method: picosecond and femtosecond pulses leave only a small residual 0 of heat, suitable to this desired effect, because of the evaporative cooling effect described above. Thus the method disclosed herein of delivering fluence has this special advantage to the material processing not available if the repetition rates are low or if the pulses are not ultrafast.
Thus, in accordance with the present invention, applying high frequency ~5 bursts (2, 3, 4, ..., pulses, through to continuous high repetition rate pulsetrains) at frequencies of few MHz to 100's of MHz provides control over the thermal physics not available with low-pulse (<1MHz) rate laser systems because thermal transport removes dissipated laser energy not carried away by the plume. The thermal heat extend is intermediate between long pulse interactions 2o and single-ultrafast (<1 MHz) laser interactions. The process is widely applicable to all classes of materials and of general advantage to the following processes: machining, micromachining, surface structuring, annealing, shock treatment, refractive index profiling, via formation, surface cleaning.
The pulse-to-pulse separation (inverse of frequency) is the key control 25 parameter this novel method that provides several significant advantages.
The sufficiently long separation between laser pulses permits hydrodynamic expansion of the plume and/or plasma to avoid laser shielding effects; the subsequent laser pulse is not reflected/absorbed by the plume and all or most of the laser energy strikes the target surface for high efficiency energy coupling.

The degree of plume/plasma shielding is controlled by the pulse-to-pulse separation (amongst other parameters such as hole depth).
Sufficiently short separation retains a thermal component in the target material that presents heated/modified material to successive ultrafast laser pulses in the burst. This thermal component is key to numerous attributes (described below) that are not available with low repetition rate (<1 MHz) ultrafast lasers. The degree of thermal component (surface temperature) is controlled by the pulse-to-pulse duration (amongst other parameters such as fluence, laser spot size). The thermal component can modify permanently the material o properties, providing refractive index changes in optical materials, densificiation or swelling of materials, visible marks.
The thermal component can anneal the surrounding material improving the overall quality of the laser process. The annealing process 'heals' a material in certain situations that, for example, eliminates incubation processes that ~5 change the absorption and other material properties, and make lower-rep rate laser processing less predictable. Surface swelling can also be avoided and precise rates of processing become available.
The thermal component raises the temperature in the surrounding processing volume, changing the state of the material to one possibly more 2o conducive to the laser process. In brittle materials, the higher temperature coverts the material to ductile, preventing and/or reducing the initiation of microcracks or other defects that can propagate by laser induced shock and other processing mechanisms. This brittle to ductile transition is particularly attractive to brittle materials such as glasses, wide bandgap materials, 25 semiconductors, ceramics, layered materials, and composites, providing the means for forming excisions without surface cracks, sheared flakes, collateral damage, defect formation and the like.
The thermal component reduces the required laser fluence (per pulse) with several advantages including more energy efficient material removal, excision of high aspect ratio holes, deeper holes are possible and the process is attractive for good conductors such as metals or semiconductors. The burst mode permits the excision of through holes in a single pulse burst in foils and thicker (~1 mm) metal plates thereby providing faster processing than lower repetition rate applications and low dwell-time advantages.
The present method reduces laser shock during processing, reducing the potential for damage (microcracking, exfoliation, shearing, delamination) since most laser energy is carried away in plume (ultrafast laser advantage), the small remaining thermal part is controllable by the pulse separation; this avoids gross o melting that is characteristic of long pulse (nanosecond or longer) laser processing while retaining the advantages of ultrafast laser processing.
Since ultrafast laser machining frequently supports material vaporization, the high repetition rate (>1 MHz) permits a controlled cooling phase of remaining material that prevents the formation of a melt phase normally present with long ~5 pulse or cw laser interactions; this provides better quality excisions with less debris and splattered melt.
Therefore, the pulse-to-pulse separation is a new control 'knob' to be tuned to an optimal value (less than1000 ns) that depends on parameters including the laser fluence, wavelength, material properties, beam area and 20 layout geometry. The key is to control the surrounding material temperature and phase (shorter increases temperature) while permitting enough time for the plume/plasma to expand and open a transparent path to the sample surface for the next pulse. The volume of heated region is controllable, and small, does not have to damage remaining material thereby minimizing collateral damage.
25 The foregoing description of the preferred embodiments of the invention has been presented to illustrate the principles of the invention and not to limit the invention to the particular embodiment illustrated. It is intended that the scope of the invention be defined by all of the embodiments encompassed within the following claims and their equivalents.

Claims (3)

1. A method of laser material processing, comprising:
providing a material to be processed and applying at least one burst of laser pulses to a target zone on the material, the laser pulses having a time separation between individual laser pulses sufficiently long to permit hydrodynamic expansion of a plume and/or plasma so that a next subsequent laser pulse is not substantially reflected, scattered and/or absorbed by the plume and/or plasma, and the laser pulses having a time separation between laser pulses sufficiently short so that a thermal component in the target zone presents heated material to successive laser pulses.

2. The method according to claim 1 wherein said laser pulses are applied at a frequency greater than about one megahertz.

3. A method of laser material processing, comprising:
providing a material to be processed and applying at least one burst of laser pulses to a target zone within the material, the laser pulses having a time separation between individual laser pulses sufficiently long to permit acoustic and thermal shock to the material to spread and/or dissipate so that a next subsequent laser pulse is not substantially reflected, scattered and/or absorbed by the temporarily altered material properties, the laser pulses having a time separation between individual laser pulses sufficiently short so that a thermal component in the target zone presents heated material to successive ultrafast laser pulses in the burst, control of which residual or accumulated heating serves the purpose of preventing or mitigating against the deleterious effects of material stresses in the material due to acoustic or thermal shock, while optimizing the useful range of such effects.