Nowadays, the relevance of ultrashort laser is well established for many medical or industrial applications. Indeed, the ultrashort laser technology has reached a high level of robustness which makes it compatible with the needs of industry. This laser technology combines the unique capacity to process any type of material with an outstanding precision and a minimal heat affected zone. Thanks to high average power and high repetition rate it is possible to achieve high throughput providing that the operating parameters are finely tuned to the application, otherwise heat accumulation and heat affected zone may appear. In this paper, the authors report on high throughput single pass ablation of stainless steel with a high average power Yb-doped fiber ultrashort pulse laser which is tunable in pulse duration from 350 fs to 10 ps and in repetition rate from 200 kHz to 2 MHz. The influence of pulse duration, repetition rate, fluence, energy dose, and scanning velocity will be discussed in terms of ablation efficiency and processing quality. These results will be compared to those previously obtained on aluminum, copper, and molybdenum. The authors will see that the effect of these parameters is strongly material dependent. The ablation behavior of stainless steel is very sensitive to these parameters meanwhile it is not the case for aluminum in the investigated process window. The authors observe an intermediate behavior for copper and molybdenum. Moreover, the authors will demonstrate that engraving of metals without melt formation is possible even at high average power (20 W) and high repetition rate. Furthermore, best ablation efficiency and quality are obtained at low fluencies. Scaling up from 2 to 15 W implies to use high repetition rate and high deflection velocity.

Today, there is an increasing interest for high throughput surface processing, especially for texturing and engraving large manufacturing tools for different industrial fields such as injection molding, embossing, and printing. These are quite hard to produce since it is necessary to have a good compromise between high removal rate and high quality issues (low roughness, burr-free, narrow heat affected zone). Short pulse lasers are commonly used for this application, but the main limitation is the uncontrolled melting of the processed area at average power (typically 50 W). Therefore, the removal rate with short pulse lasers is about 10 mm3/min. In this context, ultrashort pulse technology seems to be promising and should be able to overcome this limitation since it is possible to achieve high throughput and outstanding processing quality at the same time, providing that the operating parameters are finely tuned to the application in order to minimize the thermal load into the target material.

In the present paper, we present some results on high throughput single pass ablation of stainless steel with a high average power Yb-doped fiber ultrashort pulse laser which is tunable in pulse duration from 350 fs to 10 ps and in repetition rate from 200 kHz to 2 MHz. Several parameters of influence, such as pulse duration, repetition rate, fluence, energy dose, and scanning velocity, are considered. These results are discussed in terms of ablation efficiency and processing quality, and compared to those previously obtained on aluminum, copper, and molybdenum.

The ablation mechanism with ultrashort pulses is governed by the laser properties as well as the optical and the thermo-physical properties of the material. The ultrashort ablation mechanism has been described in several key papers in the last decades.1–4 For metals, the absorption generally occurs in free electrons of the conductive band thanks to inverse Bremsstrahlung. This absorption is followed by a fast energy relaxation and thermalization within the electronic subsystem on a femtosecond timescale, a localized heat diffusion by hot electrons, and finally an energy transfer from hot electrons to the lattice on a longer time scale, ranging from few picoseconds to few tens of picoseconds for different metals, owing to the photon-electron coupling.5 

Parameters of influence in surface ablation and their process window have been reviewed and discussed in previous papers.1,2,6–10

In the middle of the 90 s Momma et al. have published very impressive micrographies of stainless steel drilled in the nanosecond, picosecond, and femtosecond regimes, showing that ultrashort pulses enable to reduce burr and droplets in the vicinity of the hole.2 Indeed, thanks to the extremely short of the pulse, the thermal diffusion into the surrounding bulk material can be neglected. Lopez et al. have underlined that the effect of pulse duration is material dependent.7 

Basically, etch rate normally increases with fluence. Chichkov et al. have shown that in the femtosecond-picosecond range near-threshold fluences lead to a better ablation quality than higher fluences.1 So, low fluence leads to high processing quality and low removal rate (optical regime), meanwhile high fluence induces poor processing quality and removal rate (thermal regime).

In addition, Lopez and Neuenschwander have recently pointed out an optimum in fluence near the ablation threshold that gives the best etch rate per average power or etch rate per fluence.6,7,10–13 Neuenschwander et al. have also proposed a model to explain this tendency.11–13 This optimum in fluence is shifted to lower fluences for shorter pulse duration.11 The maximum value in etch rate per average power also increases for shorter pulse duration.

Vorobyev et al. have shown that heat accumulation is possible in the femtosecond regime.14 Heat accumulation enhances the removal rate but also introduces some detrimental effects on the processing quality.6,7,9 The effect of heat accumulation increases at high repetition rates9 and high fluences.1 

Furthermore, particles and plasma shielding may occur if the temporal delay between two consecutive pulses is too short with respect to the lifetime of the ablation plume. Indeed, König et al. have used a pump-probe-based time-resolved plasma attenuation experiment in order to measure the plume lifetime during laser ablation of aluminum with high repetition rate in the femtosecond regime. With their set of parameters, particles shielding occurs above 200 kHz and plasma shielding above 5 MHz.15 

Ancona et al. have measured the number of pulses to go through a stainless steel foil with picosecond and femtosecond pulses and with repetition rates ranging from 100 kHz to 1 MHz. First, the authors show that higher drilling efficiency is obtained for shorter pulses (less pulses are required). Second, rising repetition rate induces two antagonist phenomena: heat accumulation14 on one hand and particles shielding15 on the other hand. Heat accumulation improves the material removal rate to the detriment of the processing quality meanwhile particles shielding reduces the pulse energy available to produce ablation.9 In the range of 300–500 kHz, the heat accumulation induced by high repetition rate becomes sufficient enough to overcome the negative effect of particles shielding and to enhance the drilling rate.9 Third, the effect of heat accumulation is enhanced with increasing fluences. In the same paper, they have also shown that the effects of heat accumulation and particles shielding are material dependent, since they are less pronounced on copper with respect to steel.9 

All trials on stainless steel have been performed with a commercial Yb-doped fiber ultrashort laser from Amplitude Systemes (model Tangerine). The operating wavelength is 1030 nm. The maximum average power is 20 W. The M2 factor is about 1.2. The repetition rate is ranging from 200 kHz to 2 MHz. An internal pulse picker enables to generate lower repetition rates down to 1 kHz. The pulse duration can be easily tuned from 300 fs to 10 ps, but it cannot be set above 1 ps at 200 kHz. The pulse duration and the average power were measured with a Pulse Check autocorrelator and a Coherent powermeter, respectively.

The experimental setup includes the following elements: the Tangerine laser, a halfwave plate and a polarizer cube for power tuning, a 2×-beam expander, a 2-axis galvo head (Scanlab) for beam motion on the target, a 100 mm-f(theta) focusing lens, a set of XYZ-motorized stages (Newport) for focus setting and for positioning the sample under the laser beam, and finally a sample holder. The optical transmission between the laser output and the target is about 90% ± 5%. The average power measurement is made after the focusing lens. The spot size (at 1/e2) is measured thanks to a BeamMap2 M2-meter. The spot size is 28 ± 2 μm for all trials. The fluence is calculated by dividing the pulse energy by the spot area. The resulting fluence has been varied from 0.2 to 13 J/cm2 for all trials.

Trials have been done on austenitic stainless steel AISI 316L foils with a thickness of 100 μm. These samples were provided from Goodfellow. We applied no cleaning before laser treatment.

The experimental protocol is based on a single pass process. For each set of parameters, in terms of pulse duration, repetition rate, and average power, we have produced a pattern composed by 14 parallel lines with scanning velocities ranging from 1 to 2000 mm/s. Each line produces a groove on the target. Depth and width depend on the operating parameters.

The scanning velocities have been checked by measuring the distance between two subsequent spots at low repetition rate. The polarization of the laser beam is linear and perpendicular to the motion of the beam on the target. We have considered four levels in average power: 2, 5, 10, and 15 W. Thus, depending on the repetition rate and on the scanning velocity, we have collected data with fluences ranging from 0.2 to 13 J/cm2. The pulse-to-pulse linear overlap ranges from 64% to 99.9%. So, each point of the surface has received from 6 to 1100 pulses. Uncertainties are about ± 9% on fluence, ± 4% on etch rate, and ± 9% on ablation efficiency.

The results obtained on stainless steel have been compared to those previously obtained on aluminum, copper, and molybdenum using a noncommercial 40 W-tunable chirped pulse amplification Yb-doped fiber ultrashort laser made by Amplitude Systemes and an optical setup similar to the one used in the present work.7 

The profile of each groove (depth and width) was measured using a Leica DCM 3D confocal microscope based on light-emitting diode technology with a 100× /0.90-objective. Each depth measurement is the mean of five measurements. The width was ranging from 10 to 34 μm, depending on the pulse duration, the scanning velocity, and the average power. The most significant measurements were obtained on grooves with depth between 0.3 and 22 μm, and with aspect ratio below 1. For groove with aspect ratio above 1, the roughness and the shape of sidewalls induce light trapping and shadowing effects which reduce the accuracy of the measurement. Furthermore, below 0.3 μm the groove is not deep enough, compared to the initial sample roughness, to get a significant depth measurement. Since each groove has a triangularlike cross-sectional area, the etch rate per pulse or per minute is calculated from the cross-sectional area, the scanning velocity, and the repetition rate.

Scanning electronic microscopy (SEM) analyzes were also performed using a Phenom ProX microscope in order to observe the processing quality in the vicinity of the grooves. This SEM microscope offers two modes: the full mode which gives a realistic view of the surface and the topography mode which is more suitable to enhance high and low-relief. The samples have been cleaned in an ultrasonic bath (3 min, water, 30 °C) before SEM analysis.

Etch rate is the material removal rate either per pulse or per minute. The unit is (μm3·pls−1) or (mm3·min−1), respectively. Since we have tried several scanning velocities for each set of parameters (pulse duration, repetition rate and average power), we may have several exploitable grooves and consequently several operating points. We assume that the optimum scanning velocity may change from one set of parameters to another. So for each set of parameters, we will only keep the operating point leading to the highest etch rate, which is called the best operating point. This protocol has been described in detail in a previous paper.7 

From a practical point of view, is it more relevant for a given pulse energy to use a small spot with high fluence or a large spot with a low fluence? How to set the operating parameters in order to get the best process in terms of efficiency?

To answer these questions, we introduce the ablation efficiency (ρ) as the ratio between the minimum energy required to transform 1 mm3 from solid to vapor with a calculation based on thermo dynamical laws, called Eth (J·mm−3), and the energy required to engrave a 1 mm3-groove in our experiment, called Eexp (J·mm−3). The formula used for calculation of ablation efficiency is the following:

ρ=EthEexp,
(1)

where Eexp is calculated from our experimental data using the following formula:

Eexp=PavS×V,
(2)

where Pav is the average power on the target (W), S is the groove cross section (mm2), and V is the scanning velocity (mm·s−1).

Meanwhile, Eth is calculated by adding melting enthalpy (ΔHmelt), boiling enthalpy (ΔHboil), and the enthalpy required to heat the metal from ambient temperature to a temperature just above its boiling temperature (ΔH), as exhibited in the following formula:

Eth=ΔHmelt+ΔHboil+ΔH,
(3)

where ΔH is defined as

ΔH=Cp(T)×ΔTSolliq+Cp(T)×ΔTLiqGas.
(4)

For basic element, the enthalpy ΔH can be calculated using the Shomate equation.16 For alloy, the enthalpy ΔH is estimated using the enthalpy of each component weighted according to the relative composition of this alloy, which is about 69% Fe, 17% Cr, 12% Ni, and 2% Mo, in case of 316L stainless steel. All data are regrouped in Table I.

TABLE I.

Theoretical calculation of energy required to vaporize 1 mm3 of metal (Eth).

MaterialΔH (J/mm3)ΔHmelt (J/mm3)ΔHboil (J/mm3)Eth (J/mm3)
Stainless steel 18.4  59.6 80 
Aluminum 8.0 1.1 34.0 43.1 
Copper 12.9 1.7 47.6 62.2 
Molybdenum 21.6 4.5 70.6 96.7 
Iron 18.8 1.7 58.6 79.1 
Chrome 16.6 3.6 54.6 74.8 
Nickel 18.4 2.6 65.2 86.2 
MaterialΔH (J/mm3)ΔHmelt (J/mm3)ΔHboil (J/mm3)Eth (J/mm3)
Stainless steel 18.4  59.6 80 
Aluminum 8.0 1.1 34.0 43.1 
Copper 12.9 1.7 47.6 62.2 
Molybdenum 21.6 4.5 70.6 96.7 
Iron 18.8 1.7 58.6 79.1 
Chrome 16.6 3.6 54.6 74.8 
Nickel 18.4 2.6 65.2 86.2 

The ablation efficiency is a dimensionless ratio. In case of an ideal process, ablation efficiency should be one since all the incoming energy is used for ablation. However, in case of direct laser ablation process, ablation efficiency is generally less than one since part of the incoming energy is spent in particles shielding, light scattering, surface reflection, or heating. So, the ratio Eth/Eexp describes how close to an ideal process we are. The higher the ratio is the more efficient will be the engraving process.

The maximum ablation efficiency, obtained for each set of parameters, can be plotted as a function of fluence, repetition rate, pulse duration energy, or dose per millimeter, in order to determine the influence of operating parameters and their process window.

Fluence is one of the key parameters in laser surface processing which influences both the kinetics and the quality of ablation. In Fig. 1, we have plotted etch rate versus fluence using a logarithmic scale, for different pulse durations ranging from 300 fs to 10 ps. As expected the etch rate grows with fluence. The transition between optical and thermal regimes occurs nearby 1 J/cm2, in accordance with previous investigations.1,2

FIG. 1.

Etch rate (μm3·pls−1) as a function of fluence (J·cm2) for a steel target with different pulse durations ranging from 300 fs to 10 ps.

FIG. 1.

Etch rate (μm3·pls−1) as a function of fluence (J·cm2) for a steel target with different pulse durations ranging from 300 fs to 10 ps.

Close modal

Figure 2 shows the ablation efficiency dependence on fluence for different pulse durations ranging from 300 fs to 10 ps. Target material is stainless steel. The highest efficiency is obtained nearby 0.16 J/cm2 and its value is 0.45 at 300 fs, which means that more than half of the incoming pulse energy is lost in heating, particles shielding or light scattering. All the curves exhibit a linear course with a negative slope. It means that efficiency drops with increasing fluence. So, high fluence induces high throughput but also low ablation efficiency. Therefore, in terms of process efficiency, working at fluence above 1 J/cm2 is not beneficial, but for taper consideration (deep processing for instance). However, according to previous works, we expect a drop in efficiency below the optimum fluence (0.16 J/cm2).6,7,11

FIG. 2.

Ablation efficiency as a function of fluence (J·cm2) for a steel target with different pulse durations ranging from 300 fs to 10 ps.

FIG. 2.

Ablation efficiency as a function of fluence (J·cm2) for a steel target with different pulse durations ranging from 300 fs to 10 ps.

Close modal

As shown in Figs. 1 and 2, both etch rate and ablation efficiency drop with increasing pulse duration from 300 fs to 10 ps on steel.

The same experiment has been conducted on aluminum, copper, and molybdenum. The results obtained at 300/400 fs are regrouped in Fig. 3. The optimum fluence is nearby 0.16 for steel, 0.5 for aluminum and molybdenum, and finally 3 J/cm2 for copper. Therefore, the ablation efficiency is highly material dependent. We assume that the reason is the difference between metals in terms of thermo-physical properties (such as melting point, thermal conductivity, heat capacity, and electron-phonon coupling constant).

FIG. 3.

Ablation efficiency as a function of repetition rate (kHz) for stainless steel, aluminum, copper, and molybdenum. Pulse duration is ranging from 300 to 400 fs. Average power is 2 W.

FIG. 3.

Ablation efficiency as a function of repetition rate (kHz) for stainless steel, aluminum, copper, and molybdenum. Pulse duration is ranging from 300 to 400 fs. Average power is 2 W.

Close modal

A similar experiment is performed with different pulse duration ranging from 300 fs to 10 ps. Then it is possible to plot the maximum ablation efficiency as a function of pulse duration (Fig. 4). Since repetition rate has a major effect on efficiency for steel (see next paragraph), we have separately considered kHz and MHz regimes for this material. Then, we notice that pulse width has a major negative effect on steel (−60%) and molybdenum (−50%), a slight effect on copper (−30%), and no significant effect on aluminum (the variation is smaller than uncertainty). This tendency confirms results previously published by Ancona9 and Neuenschwander.10 

FIG. 4.

Ablation efficiency as a function of repetition rate (kHz) for steel aluminum, copper, and molybdenum.

FIG. 4.

Ablation efficiency as a function of repetition rate (kHz) for steel aluminum, copper, and molybdenum.

Close modal

As shown in Fig. 5, ablation efficiency on steel rises with increasing repetition rate from 200 kHz to 2 MHz. We observe a major positive effect on steel (+80%). On the other hand, we see no significant effect on aluminum. Copper and molybdenum exhibit an intermediate behavior since, the curve rises until 700 kHz to 1 MHz then there is a slight decrease in efficiency. Actually, the effect of repetition rate is highly material dependent, and probably related to thermo-physical properties of each material (melting and boiling enthalpy, heat conductivity, electron-to-phonon coupling time). For steel, we underline that average power is constant (2 W) in Fig. 5 since this material requires fluence levels as low as possible (see Fig. 3).

FIG. 5.

Ablation efficiency as a function of repetition rate (kHz) for stainless steel, aluminum, copper, and molybdenum. Pulse duration is ranging from 300 to 400 fs.

FIG. 5.

Ablation efficiency as a function of repetition rate (kHz) for stainless steel, aluminum, copper, and molybdenum. Pulse duration is ranging from 300 to 400 fs.

Close modal

In terms of processing quality, the effect of repetition rate depends on the scanning velocity and the capacity of the material to cope with the increasing thermal load. Indeed, at low velocity (250 mm/s), rising repetition rate enhances heat affected zone on steel until producing extensive burr and groove collapse (Fig. 6). On the other hand, at high velocity (2 m/s) it is possible to scale up throughput while maintaining a good processing quality (Fig. 7). Ablation efficiency is 0.07 at 200 kHz, 0.11 at 500 kHz, 0.15 at 1 MHz, and 0.18 at 2 MHz.

FIG. 6.

SEM micrographies of steel with different repetition rate ranging from 200 kHz to 2 MHz at low velocity (250 mm/s). Parameters: pulse duration 300/400 fs, average power 15 W. Fluence ranges from 13 to 1.3 J/cm2. Number of pulses per point goes from 24 to 224.

FIG. 6.

SEM micrographies of steel with different repetition rate ranging from 200 kHz to 2 MHz at low velocity (250 mm/s). Parameters: pulse duration 300/400 fs, average power 15 W. Fluence ranges from 13 to 1.3 J/cm2. Number of pulses per point goes from 24 to 224.

Close modal
FIG. 7.

SEM micrographies of steel with different repetition rate ranging from 200 kHz to 2 MHz at high velocity (2 m/s). Parameters: pulse duration 300/400 fs, average power 15 W. Fluence ranges from 13 to 1.3 J/cm2. Number of pulses per point goes from 3 to 28.

FIG. 7.

SEM micrographies of steel with different repetition rate ranging from 200 kHz to 2 MHz at high velocity (2 m/s). Parameters: pulse duration 300/400 fs, average power 15 W. Fluence ranges from 13 to 1.3 J/cm2. Number of pulses per point goes from 3 to 28.

Close modal

It is possible to establish a correlation between the amount of energy deposited per millimeter and the ablation efficiency on one side, or the processing quality on the other side? To answer this question, we define dose per millimeter (δ) as the following ratio:

δ=v×EV,
(5)

where v is the repetition rate (s−1), E is the energy per pulse (J), and V is the scanning velocity (mm·s−1). The unit for dose is J·mm−1.

In Fig. 8, we have plotted ablation efficiency as a function of dose for a steel target and a pulse duration ranging from 300 to 400 fs. We define the critical dose as the dose beyond which there is onset of melting and then drop in ablation efficiency. Indeed, above the critical dose, a significant part of the incoming laser energy is spent in heating and is not available to produce ablation, so the ablation process is less efficient. We also define the maximum dose as the dose beyond which the groove starts to collapse due to melting and there is no engraving anymore. So, the maximum dose described the capability of the target material to cope with the heat accumulation. Both critical and maximum doses are experimental data and extracted from the Fig. 8. Therefore, the Fig. 8 can be used as an abacus. It is possible to maintain a good processing quality and process efficiency as long as the actual dose is below the critical dose. For instance, with a 15 W-average power and a 200 kHz-repetition rate, the minimum scanning velocity is 250 mm/s (see formula 5). So it is possible to maintain quite good ablation efficiency on a wide process window, in terms of dose, repetition rate, fluence, or velocity, providing that the other parameters are finely tuned.

FIG. 8.

Ablation efficiency as a function of energy dose per millimeter (J mm−l) for a steel target with pulse duration ranging from 300 to 400 fs. The black line represents spline fit.

FIG. 8.

Ablation efficiency as a function of energy dose per millimeter (J mm−l) for a steel target with pulse duration ranging from 300 to 400 fs. The black line represents spline fit.

Close modal

Therefore, we have identified three regimes. The first one leads to high efficiency and processing quality (below critical dose). The second one is an intermediate regime with a lower efficiency and quality due to melting (between critical and maximum dose). The third one is the inefficient regime which corresponds to the groove collapse (above maximum dose). As shown in Fig. 9, these three regimes can be correlated to SEM observations.

FIG. 9.

SEM micrographies of steel engraved in efficient regime (upper left, dose 0.005 J/mm and efficiency 0.38), in intermediate regime (upper right, dose 0.04 J/mm and efficiency 0.21) and in inefficient regime (bottom, dose 0.064 J/mm and zero-efficiency) with pulse duration ranging from 300 to 400 fs.

FIG. 9.

SEM micrographies of steel engraved in efficient regime (upper left, dose 0.005 J/mm and efficiency 0.38), in intermediate regime (upper right, dose 0.04 J/mm and efficiency 0.21) and in inefficient regime (bottom, dose 0.064 J/mm and zero-efficiency) with pulse duration ranging from 300 to 400 fs.

Close modal

As shown in Fig. 10, although ablation efficiency drops with rising pulse duration from 300 fs to 10 ps, pulse duration has no significant effect on both critical and maximum dose values.

FIG. 10.

Ablation efficiency (no unit) as a function of energy dose per millimeter (J·mm−1) for a steel target with different pulse duration ranging from 300 fs to 10 ps. Color lines represent spline fit.

FIG. 10.

Ablation efficiency (no unit) as a function of energy dose per millimeter (J·mm−1) for a steel target with different pulse duration ranging from 300 fs to 10 ps. Color lines represent spline fit.

Close modal

The same experiment has been conducted on aluminum, copper, and molybdenum. The results are regrouped in Fig. 11 and the experimental values of critical and maximum doses are shown in Table II. Due to its high thermo-physical properties molybdenum can stand a high thermal load (up to 0.4 J/mm). On the other hand, stainless steel and aluminum are very sensitive to thermal load (critical dose below 0.1 J/mm). Copper exhibits an intermediate behavior (critical dose is 0.2 J/mm). Moreover, the groove produced on copper collapses above 0.22 J/cm2 (no exploitable groove above this value).7 

FIG. 11.

Ablation efficiency as a function of energy dose per millimeter (J mm−l) for steel, aluminum, copper, and molybdenum, with pulse duration ranging from 300 to 400 fs.

FIG. 11.

Ablation efficiency as a function of energy dose per millimeter (J mm−l) for steel, aluminum, copper, and molybdenum, with pulse duration ranging from 300 to 400 fs.

Close modal
TABLE II.

Critical and maximum energy dose per mm for various metals.

MaterialCritical dose (J/mm)Max. dose (J/mm)
Stainless steel 0.02 0.06 
Aluminum 0.08 0.13 
Copper 0.22 <0.3 
Molybdenum 0.22 0.43 
MaterialCritical dose (J/mm)Max. dose (J/mm)
Stainless steel 0.02 0.06 
Aluminum 0.08 0.13 
Copper 0.22 <0.3 
Molybdenum 0.22 0.43 

As shown in Fig. 12, thermal load rises with decreasing scanning velocity. Table III presents the number of pulse per point and the ablation efficiency corresponding to the four micrographies presented in Fig. 12. We notice that heat accumulation improves the ablation efficiency but it also enhances thermal detrimental effects, until producing the groove collapse.

FIG. 12.

SEM micrographies of steel with different scanning velocities ranging from 2000 to 250 mm/s. Operating parameters are: pulse duration 400 fs, rep. rate 2 MHz, average power 16 W, fluence 1.3 J/cm2.

FIG. 12.

SEM micrographies of steel with different scanning velocities ranging from 2000 to 250 mm/s. Operating parameters are: pulse duration 400 fs, rep. rate 2 MHz, average power 16 W, fluence 1.3 J/cm2.

Close modal
TABLE III.

Ablation efficiency vs scanning velocity on steel (400 fs, 2 MHz, 16 W, and 1.3 J/cm2).

Scan. velocity (mm/s)Number of pulses per pointAblation efficiency
2000 28 0.18 
750 75 0.17 
500 112 0.21 
250 225 
Scan. velocity (mm/s)Number of pulses per pointAblation efficiency
2000 28 0.18 
750 75 0.17 
500 112 0.21 
250 225 

From a practical point of view what could be the processing time to remove 1 mm3 of stainless steel with an ultrashort laser? Tables IV and V show the time to remove 1 mm3 of stainless steel at 15 and 2 W, respectively. These values are extracted from our experiment database. We compared 300/400 fs to 10 ps on one hand, and 200 kHz to 2 MHz on the other hand. The shorter processing time is obtained for shorter pulse width and higher repetition rate at 2 m/s (400 fs and 2 MHz). It takes 28 s at 15 W and 89 s at 2 W to ablate 1 mm3 of steel. Corresponding SEM pictures are regrouped in Fig. 13. The best processing quality is obtained at low fluence.

TABLE IV.

Time required to remove 1 mm3 of stainless steel at 15 W.

Material300 fs 200 kHz400 fs 2 MHz10 ps 2 MHz
Fluence (J/cm213 1.3 1.3 
Efficiency 0.08 0.18 0.11 
Time to remove 1 mm3 (s) 66 28 45 
Material300 fs 200 kHz400 fs 2 MHz10 ps 2 MHz
Fluence (J/cm213 1.3 1.3 
Efficiency 0.08 0.18 0.11 
Time to remove 1 mm3 (s) 66 28 45 
TABLE V.

Time required to remove 1 mm3 of Stainless steel at 2 W.

Material300 fs 200 kHz400 fs 2 MHz10 ps 2 MHz
Fluence (J/cm21.6 0.16 0.16 
Efficiency 0.25 0.45 0.15 
Time to remove 1 mm3 (s) 162 89 260 
Material300 fs 200 kHz400 fs 2 MHz10 ps 2 MHz
Fluence (J/cm21.6 0.16 0.16 
Efficiency 0.25 0.45 0.15 
Time to remove 1 mm3 (s) 162 89 260 
FIG. 13.

SEM micrographies of steel engraved at 2 W (left) and 15 W (right), with the set of parameters leading to the best ablation efficiency. Pulse duration is 400 fs. Repetition rate is 2 MHz. Velocity is 2 m/s.

FIG. 13.

SEM micrographies of steel engraved at 2 W (left) and 15 W (right), with the set of parameters leading to the best ablation efficiency. Pulse duration is 400 fs. Repetition rate is 2 MHz. Velocity is 2 m/s.

Close modal

So, increasing the average power by a factor of 7.5 improves the processing time by only a factor of 3.2. The up-scaling is not linear since ablation efficiency drops with increasing fluence. Therefore, high throughput requires combining high average power, while maintaining low fluence and low pulse-to-pulse overlap.

We have performed high throughput single pass ablation of stainless steel with a high average power and high repetition rate ultrashort pulse laser, with good ablation efficiency (up to 0.45). The effect of fluence, pulse duration, repetition rate, dose, and scanning velocity are highly material dependent. Ablation of stainless steel is very sensitive to these parameters. Moreover, the best ablation efficiency and quality are obtained at low fluences (<0.2 J/cm2). So, scaling up the average power from 2 to 15 W implies to use high repetition rate and high deflection velocity. Concerning this latter point, there is a need for new scanning technology beyond 2 m/s. Furthermore, deep engraving will require multipass processing.

The authors acknowledge the European Commission, the French Ministry of Research, and the Aquitaine Regional Council for support and funding.

1.
B. N.
Chichkov
,
C.
Momma
,
S.
Nolte
,
F.
Von Alvensleben
, and
A.
Tünnermann
, “
Femtosecond, picosecond and nanosecond laser ablation of solids
,”
Appl. Phys. A
63
,
109
115
(
1996
).
2.
C.
Momma
,
B. N.
Chichkov
,
S.
Nolte
,
F.
Von Alvensleben
,
A.
Tünnermann
,
H.
Welling
, and
B.
Wellegehausen
, “
Short-pulse laser ablation of solid targets
,”
Opt. Commun.
129
,
134
142
(
1996
).
3.
S.-S.
Wellershoff
,
J.
Hohlfeld
,
J.
Güdde
, and
E.
Matthias
, “
The role of electron-phonon coupling in femtosecond laser damage of metals
,”
Appl. Phys. A
69
,
S99
S107
(
1999
).
4.
J.
Byskov-Nielsen
,
J.-M.
Savolainen
,
M. S.
Christensen
, and
P.
Balling
, “
Ultra-short pulse laser ablation of copper, silver and tungsten: experimental data and two-temperature model simulations
,”
Appl. Phys. A
103
,
447
453
(
2011
).
5.
S. I.
Anisimov
and
b. S.
Lukyanchuk
, “
Selected problems of laser ablation theory
,”
Phys. Usp.
45
,
293
324
(
2002
).
6.
J.
Lopez
,
R.
Torres
,
Y.
Zaouter
,
P.
Georges
,
M.
Hanna
,
E.
Mottay
, and
R.
Kling
, “
Study on the influence of repetition rate and pulse duration on ablation efficiency using a new generation of high power Ytterbium doped fiber ultrafast laser
,”
Proc. SPIE
8611
,
861118
(
2013
).
7.
J.
Lopez
,
Y.
Zaouter
,
R.
Torres
,
M.
Faucon
,
C.
Hönninger
,
P.
Georges
,
M.
Hanna
,
E.
Mottay
, and
R.
Kling
, “
Parameters of influence in surface ablation of metals with using a high power tunable ultrafast laser
,” in
Proceedings of the International Congress on Applications of Laser & Electro-Optics, M303, 2013
, pp.
686
694
.
8.
R.
Le Harzic
,
D.
Breitling
,
M.
Weikert
,
S.
Sommer
,
C.
Föhl
,
S.
Valette
,
C.
Donnet
,
C. E.
Audouard
, and
F.
Dausinger
, “
Pulse width and energy influence on laser micromachining of metals in a range of 100 fs to 5 ps
,”
Appl. Surf. Sci.
,
249
,
322
331
(
2005
).
9.
A.
Ancona
,
S.
Döring
,
C.
Jauregui
,
F.
Röser
,
J.
Limpert
,
S.
Nolte
, and
A.
Tünnermann
, “
Femtosecond and picosecond laser drilling of metals at high repetition rates and average powers
,”
Opt. Lett.
34
,
3304
3306
(
2009
).
10.
B.
Neuenschwander
,
B.
Jaeggi
, and
M.
Schmid
, “
From ps to fs: dependence of the material removal rate and the surface quality on the pulse duration for metals, semiconductors and oxides
,” in
Proceedings of the International Congress on Applications of Laser & Electro-Optics, M1004, 2013
, pp.
959
968
.
11.
B.
Neuenschwander
,
B.
Jaeggi
, and
M.
Schmid
, “
From fs to sub-ns: Dependence of the material removal rate on the pulse duration on metals
,”
Phys. Procedia
41
,
794
801
(
2013
).
12.
B.
Lauer
,
B.
Neuenschwander
,
B.
Jaeggi
, and
M.
Schmid
, “
From fs-ns: Influence of the pulse duration onto the material removal rate and machining quality for metals
,” in
Proceedings of the International Congress on Applications of Laser & Electro-Optics
,
2013
, Paper No. M309.
13.
B.
Jaeggi
,
B.
Neuenschwander
,
M.
Zimmermann
,
L.
Penning
,
R.
de Loor
,
K.
Weingarten
, and
A.
Oehler
, “
High throughout and high precision laser micromachining with ps-pulses in synchronized mode with a fast polygon line scanner
,”
Proc. SPIE
8967
,
25
(
2014
).
14.
A. Y.
Vorobyev
and
C.
Guo
, “
Direct observation of enhanced residual thermal energy coupling to solids in femtosecond laser ablation
,”
Appl. Phys. Lett.
86
,
011916
(
2005
).
15.
J.
König
,
S.
Nolte
, and
A.
Tünnermann
, “
Plasma evolution during metal ablation with ultrashort pulses
,”
Opt. Express
13
,
10597
10607
(
2005
).
16.
M. W.
Chase
, Jr.
,
NIST - JANAF Thermochemical Tables, Fourth Edition, Monograph 9
(American Chemical Society and the American Institute of Physics for the National Institute of Standards and Technology,
1998
), pp.
1
-
1951
.

John Lopez is an engineer from the French graduate school of Physics and Chemistry of Bordeaux. He got his Ph.D. on laser ablation of polymers at the University of Bordeaux in 1997. He has a permanent position as a research engineer in the French CNRS since 2001. He is currently working in a public institute in the University of Bordeaux. He has written over 40 papers on laser processing in peer-reviewed papers, in international conference proceedings, and in several technology magazines in Europe. He is also the President of the French association of industrial laser users called CLP since 2010.