Open Access
Add to BibSonomy
Abstract
We report for the first time to our knowledge on top-down percussion drilling of high-quality deep holes in different glasses with femtosecond laser pulses in GHz-burst mode. We reveal the dynamics of the percussion drilling process by pump-probe shadowgraphy and thermal camera imaging demonstrating that the drilling process in GHz-burst mode is fundamentally different from single-pulse processing and confirming the presence of thermal accumulation. Moreover, we show a comparison to drilling by femtosecond single-pulses containing an equal laser fluence in sodalime, alkali-free alumina-borosilicate, fused silica, and sapphire.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Femtosecond (fs) laser micromachining in GHz-burst mode has attracted much interest in the last few years. Specific ablation rates exceeding the ones of single pulse ablation have been reported [1–5], and others showing less ablation efficiency [6,7] depending on the laser parameters [8], notably on the laser fluence and the number of pulses within the burst. However, most of the studies so far have been carried out on metals and silicon, and very few studies have been reported on dielectrics [9–11] which are focused on milling ablation removal rates. All three studies use only few pulses within the burst and show an increase in the ablation removal rate, even of a factor of 6 in sapphire [9] when passing from 1 to 25 pulses. Similar observations have been made on fused silica [10] by increasing the number of pulses from 1 to 10. However, the rise in ablation removal rate was on the expense of an increasing surface roughness. An early study using a femtosecond Ti:Sapphire laser in MHz-burst mode demonstrates very interesting results of drilling in glasses including time-resolved side-view imaging of the hole formation [12]. Indeed, in situ microscopy is a powerful tool for process control and optimization as reported for front and rear side ablation [13]. A recent review article suggests an interesting potential for GHz-burst micromachining of dielectrics [14].
In this contribution, we demonstrate for the first time to our knowledge top-down percussion drilling of dielectrics with GHz-bursts. We especially investigate the interaction dynamics of the laser beam in GHz-burst mode with the glass during the drilling process by time-resolved shadowgraphy and thermal imaging.
2. Experimental setup
2.1 Pump-probe setup for time-resolved shadowgraphy
We used an Yb-doped femtosecond laser system based on a commercial laser system (Tangor 100, Amplitude) emitting 500-fs pulses at 1030 nm, which can be operated in single pulse mode or in GHz-burst mode and is thoroughly described in [5]. For GHz-burst mode, the intra-burst repetition rate was set to 1 GHz containing 50 pulses within the burst resulting in a burst length of 50 ns, and the inter-burst repetition rate was variable between 5 Hz and 200 kHz. A microscope objective (Mitutoyo NIR APO 5x, NA = 0.14) was used to focalize the laser beam on the front surface of the different glass samples. We measured a spot diameter of 7.3 µm at 1/e2 using a CCD camera (WinCamD LCM4) and a home-made magnification system calibrated with a standard micro rule (Mitutoyo, model 375-056). Since we cover the whole entrance pupil of the objective with a collimated beam, the effective numerical aperture is 0.14 and the Rayleigh length values 41 µm. For all experiments, this laser source constitutes the pump in the pump-probe setup which is schematically shown in Fig. 1.
Fig. 1. Schematic of the pump-probe setup with two synchronized lasers for time-resolved shadowgraphy of the femtosecond GHz-burst interaction process with the glass samples.
The probe beam is delivered by an independent second laser source (Yuja, Amplitude) delivering 500-fs pulses at a wavelength of 1030 nm and 8 W average output power. This laser source was synchronized to the pump laser source by the same TTL signal from the controller of the translation stages. The probe beam passes two delay lines, a first variable one allowing for a path variation up to 2.4 m corresponding to 0–8 ns time delay thanks to motorized translation stages. A secondary optional delay line with multiple reflections adding 2.0 m per reflection offers additional path lengths from 2.5 m (1 reflection, 8.25 ns) to 14.5 m (7 reflections, 47.85 ns). The 1030-nm probe signal is then frequency doubled using a type I BBO crystal and transmitted through the sample perpendicularly to the pump beam. The probe signal is collected after two band-pass filters for 515 nm ± 10 nm by a long-distance microscope (InfiniMax KX, with MX-5 objective) followed by a second optional zoom objective (InfiniMax, x2) and imaged on a CMOS camera (Basler acA1920-25mu, 1/3.7” sensor, resolution of 1920 × 1080p, pixel size 2.2 µm x 2.2 µm, 25i/s, rolling shutter).
The camera is triggered by a TTL signal from the acousto-optic modulator at the exit of the probe laser source which is filtered and optionally delayed by an electronic generator (Tombak, Aerodiode). In practice, we focus the laser beam within the bulk 200 µm below the front surface in order to generate a permanent bulk modification by one single burst. The trigger signal generated by probe pulse N allows for imaging the bulk modification induced by pump burst or pulse N+1, respectively, whereas the induced bulk modification of pump burst or pulse N remains visible. We define the zero delay at the moment where the bulk modification induced by the pump pulse N+1 appears. The temporal resolution is determined by the probe pulse duration (500 fs) and the precision in the path delay in our setup, which is 80 µm and corresponds to 264 fs. The recording time is limited by the acquisition time of the camera (40 ms) and the buffer refreshing time (160 ms), and thus restricted to a rate of 5 Hz (200 ms).
2.2 Thermal imaging
We used the same pump laser source for drilling experiments in single-pulse and GHz-burst mode for sideview 2D thermal imaging during the drilling process. We installed a thermal camera (FLIR SC7000 InSb, spectral range 2.0–5.5 µm, pixel size 30 µm, 50 to 1000 fps) with a Zoom objective (Micro, FLIR) perpendicular to the laser beam that allows for collecting and counting the number of infrared photons. The camera was manually triggered by an interface using the software ALTAIR in order to observe the temperature evolution during the percussion drilling process in the region of interest (ROI). Note that the spatial resolution of this camera does not allow for directly observing the temperature evolution of a single hole during the drilling process as the hole diameters are too small (10–30 µm). The software allows for exporting video files in mp4 format delivering the whole information on the temperature variations during the drilling process in the ROI.
3. Results and discussion
3.1 Pump-probe shadowgraphy of single-pulse and GHz-burst percussion drilling
A first result of time-resolved shadowgraphy of the drilling process in single-pulse mode in sodalime is shown on Fig. 2. The shadowgraphs were taken with the above described pump-probe setup at a single-pulse repetition rate that was therefore voluntarily set to 5 Hz, and the delay times are zero (on the top) and 4.5 ns (on the bottom), respectively. The fluence was 289 J/cm2 per pulse, and the total drilling time was 40 s corresponding to a total of 200 single pulses. In these experiments, we used a microscope objective (Mitutoyo NIR APO 10x, NA = 0.26) leading to a spot diameter of about 3.8 µm at 1/e2 on the front surface in order to better visualize the transient effects.
Fig. 2. Shadowgraphs taken with the pump-probe setup of single-pulse drilling in sodalime for zero delay (top) and 4.5 ns delay (bottom) with a single-pulse fluence of 289 J/cm2 and a drilling time from 0 to 40 s (from the left to the right), repetition rate 5 Hz.
In single-pulse drilling, we observe a plume during the drilling process and the formation of a transient small crack at zero delay (indicated by yellow arrows). The resulting hole has a diameter of about 7 µm and a depth of 20 µm. For the delay of 4.5 ns, the images visualize the ablation plume corresponding to the ejected matter in the air counter propagating to the incident laser beam, and a wave propagating in the drilling direction within the sample is observed.
In our experiments with GHz-bursts, the number of pulses per burst was set to 50, and we applied 1000 bursts at an inter-burst repetition rate 5 Hz for pump-probe recording, resulting in a total drilling time of 200 s. Figure 3 shows the corresponding series of shadowgraphs at zero delay with burst fluences of 66 J/cm2 (top) and 109 J/cm2 (bottom), respectively, meaning that the fluences of an individual pulse are 1.3 J/cm2 and 2.2 J/cm2, respectively, which are both well below the ablation threshold of 2.9 J/cm2 reported for single-pulse ablation [15].
Fig. 3. Shadowgraphs of GHz-burst drilling in sodalime taken at zero delay time, burst fluence of 66 J/cm2(top) and 109 J/cm2(bottom), and drilling times from 0 to 200 s (from the left to the right) with 1000 bursts at an inter-burst repetition rate of 5 Hz.
In GHz-burst mode, we observe that percussing drilling takes place in a progressive way, and even at these low fluences. This is an analogous behavior like in ablation studies that have been carried out in GHz-burst mode on metals with pulse fluences below the ablation threshold. This phenomenon might be explained by an accumulative effect based on pre-heating by the first pulses of the GHz-burst analogously to the observations for ablation of metals and silicon with GHz-bursts [3–5]. For the burst fluence of 66 J/cm2, the resulting hole has a diameter of 10 µm and a depth of 205 µm which is commensurate to an aspect ratio (AR) as high as 20.
In order to visualize the progressive dynamics of the drilling process, we moved the delay line introducing a time delay of 5.6 ns and performed drilling experiments for burst fluences of 109, 219 and 328 J/cm2, respectively. Again, the number of pulses per burst was set to 50 at an inter-burst repetition rate of 5 Hz, and the drilling time was 200 s in sodalime. The corresponding series of shadowgraphs are depicted in Fig. 4.
Fig. 4. Shadowgraphs taken with the pump-probe setup of GHz-burst drilling in sodalime for burst fluences of 109, 219 and 328 J/cm2 (from the top to the bottom), respectively, and for drilling times from 0 to 200 s (from the left to the right), and taken at 5.6 ns delay time with 1000 bursts at an inter-burst repetition rate of 5 Hz.
For all parameter settings in GHz-burst mode, no transient effects appear in the sample during processing. But we observe on some shadowgraphs an ablation plume but clearly not on all of them meaning that the ejection of matter takes place in a discontinuous way. This is a fundamentally different behavior from single-pulse percussion drilling where we observe that the matter is continuously ejected (see Fig. 2), and where, of course, drilling takes place only above the single-pulse ablation threshold. We verified this behavior in GHz-burst mode for further delay times up to 48 ns on sodalime and up to 5.6 ns for other glass materials, which are not shown here since the shadowgraphs confirm the discontinuous ejection of matter and do not give any additional information. Unsurprisingly, the hole diameters enlarge with increasing burst fluence and measure 12.5, 17.5, and 20 µm for burst fluences of 109, 219 and 328 J/cm2, respectively. The corresponding hole depths can be retrieved from the images at the end of the drilling time (right column of images in Fig. 4) and measure 205, 310, and 260 µm, respectively. Then the drilling depth saturates and even decreases for higher burst fluences. However, the AR are still very high in the range of 13 to 18. The quality of the holes is discussed in section 3.3 where post-mortem microscope images are presented.
3.2 Thermal imaging
With the thermal camera described in section 2.2, we recorded the temperature variations on a wide region around the drilled holes, and especially in a region of interest (ROI) just below the front surface at the spot location during the drilling process. We carried out these experiments in sodalime and extracted images of the mp4 video file. We performed these experiments in GHz-burst mode and in single-pulse mode. First, with pulse fluence values below the ablation threshold, and second, with pulse fluence values above the ablation threshold for single-pulses. In the case of GHz-burst mode, we applied – as before – 1000 bursts with 50 pulses per burst, at an inter-burst repetition rate of 1 kHz during 1 s at burst fluences of 109 J/cm2 and 300 J/cm2, respectively, which corresponds to 50,000 pulses of 2.2 J/cm2 and 6.0 J/cm2, respectively. The latter value is well above the single-pulse ablation threshold [15]. The results are shown in Fig. 5(a) and 5(b), respectively, where the upper graph shows the temporal evolution of the temperature in the ROI on a digital level scale corresponding to single photon counts. We marked the points from 1 to 8 in the graphs where we extracted the thermal images that are shown on the lower parts of Fig. 5. In the upper right corner of the graphs, we added a microscope image of the drilled hole. The integration time of the camera was 300 µs and 1000 frames per second in a), and 200 µs and 1000 frames per second in b).
Fig. 5. Temperature evolution at the zone of interest (marked by a white star on the images 1–8) during GHz-burst irradiation of sodalime, a) below the ablation threshold for a burst fluence of 109 J/cm2 with 1,000 bursts delivered in 1 s at an inter-burst repetition rate of 1 kHz corresponding to 50,000 pulses with a fluence of 2.2 J/cm2 each, and b) above the ablation threshold for a burst fluence of 300 J/cm2 with 1,000 bursts delivered in 1 s at an inter-burst repetition rate of 1 kHz corresponding to 50,000 pulses with a fluence of 6.0 J/cm2 each.
In both cases, we observe cycles of heating and ablation, where the temperature in the ROI decreases abruptly at the moments when the matter is leaving the sample in bunches, and a residual heat increase with the typical form and saturation behavior of thermal accumulation [16]. Note, that the digital level scales corresponding to photon counts are relative, and therefore, a direct comparison of the absolute temperatures in cases a) and b) is not possible. Moreover, the color scale is dynamic during recording and only allows for visualizing temperature differences within one single image. Therefore, no color bar scale is indicated.
Interestingly, we observe 26 bunches of matter ejected during the drilling time in the case of a burst fluence of 109 J/cm2 (Fig. 5(a)), and only 3 main ejections in the case of a burst fluence of 300 J/cm2 (Fig. 5(b)). This might be due to stronger plasma shielding at higher fluences that demand more, and thus longer, thermal accumulation before the eruption of matter can take place. Further experimental and theoretical studies would be necessary to fully understand the dynamics of the matter ejection. Nevertheless, these results confirm our observations made by pump-probe shadowgraphy that the matter is leaving discontinuously. Moreover, we confirm that a part of the heat is ejected with the ablated matter as mentioned in [1–5] for the ablation of metals and semiconductor materials.
We repeated the experiment for single-pulse drilling with the same dose applied on the sample, which means 50,000 single pulses at a fluence of 2.2 J/cm2 delivered in 1s at a repetition rate of 50 kHz in the case below the ablation threshold, and 1000 single pulses at a fluence of 291 J/cm2 delivered in 1s at a repetition rate of 1 kHz in the case above the ablation threshold. The corresponding results are depicted in Fig. 6. The integration time was 1 ms at a rate of 500 frames per second in Fig. 6(a), and it was 100 µs at a rate of 1000 frames per second in Fig. 6(b)).
Fig. 6. Temperature evolution at the zone of interest (marked by a white star on the images 1–8) during single-pulse irradiation of sodalime, a) below the ablation threshold for 50,000 pulses at a single-pulse fluence of 2.2 J/cm2 delivered in 1 s at a repetition rate of 50 kHz, and b) above the ablation threshold for 1,000 pulses at a single-pulse fluence of 291 J/cm2 delivered in 1 s at a repetition rate of 1 kHz.
Unsurprisingly, there is no hole drilled in the case of single pulses with a fluence below the ablation threshold, and the temperature in the ROI increases only very moderately and saturates as the time for heat diffusion is sufficient between two consecutive pulses at a repetition rate of 50 kHz. In the case of single-pulse drilling (Fig. 6(b)), the process is continuous, a slight and unique eruption occurs during the laser irradiation, while the heat is diffusing into the sample as expected. On the microscope image in the upper right corner, we observe that drilling results in a hole with a different, more conical shape than in GHz-burst mode confirming that the obtained morphologies are different.
Moreover, sideview thermal imaging allows for adjusting and limiting the laser irradiation time when drilling a series of holes along a line in order to keep the thermal charge constant and in that way maintaining the hole depth constant. In this way, it can constitute a powerful tool for process control in real time.
3.3 Influence of the inter-burst repetition rate and drilling time
We further studied the influence of the inter-burst repetition rate, which is directly correlated with the drilling time when keeping constant the applied dose of 1000 bursts. The burst fluence was set to 112 J/cm2 at an intra-burst repetition rate of 1 GHz, and the delay was zero. We varied the inter-burst repetition rate from 5 Hz to 200 kHz, meaning that the drilling time varied between 200 s and 5 ms. The microscope images of the resulting holes in sodalime are shown on Fig. 7.
Fig. 7. Microscope images of percussion drilled holes in GHz-burst mode in sodalime with variable repetition inter-burst rates from 5 Hz to 200 kHz. The drilling time was adjusted to deliver the same dose of 1000 bursts with 50 pulses per burst at a burst fluence of 112 J/cm2.
We observe a growth in the hole depth from 5 Hz to 20 kHz. However, for inter-burst repetition rates exceeding 25 kHz, we observe an increase in the hole diameter whereas the depth diminishes. Moreover, a cone of the heat affected zone (HAZ) appears, whose base is situated at the surface of the glass sample. For repetition rates above 100 kHz, the glass is deformed by its own weight during the drilling process, and finally the hole is refilled about 50 µm below the surface. This morphology change might be due to a heat accumulation phenomenon as the time shortens between the consecutive bursts. This means that the heat accumulation is beneficial up to 20 kHz, and then turns into detrimental provoking a change in the geometry of the hole up to a collapse of the hole when reaching the softening temperature (about 700 ° C for sodalime). The appearance of the HAZ indicates that the drilling time cannot be reduced below 40 ms, corresponding to a limit in the drilling rate of 25 holes per second in our experimental conditions. Nevertheless, we are able to drill holes of 16–17 µm diameter and a depth of 340 µm corresponding to an AR of as high as 20 to 21.
3.4 Comparison of single-pulse and GHz-burst drilling
Moreover, we performed a series of experiments applying the same laser fluence during the processing, either by 1000 single pulses (pulse fluence 289 J/cm2) or distributed on 1000 GHz-bursts containing 50 pulses per burst (burst fluence 300 J/cm2) at varying single-pulse or inter-burst repetition rates of 10, 50, 100, and 200 kHz, respectively, corresponding to drilling times from 200 s to 5 ms. The results are depicted in Fig. 8 for sodalime, sapphire, alkali-free alumina-borosilicate (AF32), and fused silica.
Fig. 8. Microscope images of drilled holes for four different repetition rates (10 kHz, 50 kHz, 100 kHz, and 200 kHz) in four different materials (sodalime, sapphire, AF32, and fused silica). The images on the left side correspond to single pulse drilling and on the right side to GHz-burst drilling with 50 pulses at an intra-burst repetition rate of 1 GHz, respectively. The drilling time corresponds to 1000 single pulses (fluence 289 J/cm2 per pulse) or 1000 GHz-bursts (fluence 300 J/cm2 per burst), respectively.
We observe globally the same trend in sodalime, AF32 and sapphire. Due to the high fluence, the glasses are excessively cracked around the holes in both single-pulse and GHz-burst mode for repetition rates of 200 kHz in sodalime and sapphire. In AF32 bubbles appear at 100 and 200 kHz which are a sign of decomposition of the glass due to thermal load. Fused silica shows a better resistance to heating provoked at high inter-burst repetition rates with no effect on the drilling quality between 100 and 200 kHz. This is in agreement with our observation that fused silica is the glass with the highest ablation threshold in GHz-burst regime.
Note that single-pulse and GHz-burst percussion drilling lead to substantially different hole morphologies. In single-pulse drilling, we obtain conical holes with a rough internal surface, whereas in GHz-burst mode the holes are quasi cylindrical with a smooth internal surface. The detrimental heat accumulation effects for increasing repetition rates are clearly limiting the drilling rates that can be applied in single-pulse and GHz-burst mode drilling, respectively. However, in GHz-burst mode, drilling rates of tens to hundreds of holes per second are accessible depending on the fluence and the drilling depth that may exceed 300 µm.
3.5 Towards applications
The drilling time, which is directly linked to the burst repetition rate and limited as shown above, is an important parameter for potential applications. Therefore, we studied drilling at a fixed inter-burst repetition rate with an increasing number of bursts and thus increasing drilling time. Figure 9 shows a microscope image of drilled holes in sodalime for different drilling times from 10 ms to 100 ms (from the left to the right) in GHz-burst mode corresponding to a burst number from 100 to 1000 in steps of 100 bursts, with all bursts containing 50 pulses, at a burst fluence of 136 J/cm2 and an inter-burst repetition rate of 10 kHz.
Fig. 9. Microscope image of drilled holes in sodalime. The drilling parameters were 1000 bursts containing 50 pulses, burst fluence 136 J/cm2, inter-burst repetition rate 10 kHz for drilling times from 10 ms to 100 ms in steps of 10 ms (from the left to the right). The spot size was 9.3 µm and the effective NA = 0.10. The red dashed line indicates the front surface of the sample.
The resulting hole depths increase linearly from 73 µm (10 ms, 100 bursts) to 294 µm (100 ms, 1000 bursts), whereas the hole diameters are all similar in the range of 25–30 µm, corresponding to aspect ratios of up to 10. For an increasing number of bursts, the hole depths show a saturation behavior. All holes show a very good quality with a smooth inner surface and a hole diameter which stays constant over a big part of the depth and an even cone at the tip.
Due to the thermal character of the interaction process, care has to be taken when a matrix of holes has to be drilled in GHz-burst mode to avoid overlapping between the HAZ surrounding each hole as detrimental cooperative effects may appear if the pitch is too small compared to the HAZ. We studied percussion drilling of a matrix of through holes in a fused silica sample of 75 µm thickness. The holes were drilled from top-down and in step and repeat mode which means that one hole is drilled after another. The laser parameters were set to an inter-burst repetition rate of 100 kHz with 50 pulses per burst and a burst fluence of 219 J/cm2. The drilling time was 10 ms per hole (1000 bursts), and the distance between two holes is 100 µm. Figure 10 depicts a cutout with 16 holes of the resulting hole matrix from top view (inlet, left image) and from bottom view (outlet, right image). The obtained holes have a diameter of 24 µm at the inlet and of 21 µm at the outlet, which corresponds to a taper half angle of 1.1°, which is a very small value. At a distance of 100 µm between the holes, we do not observe any detrimental cooperative effect nor chipping. In contrary, the drilling process is very reproducible and delivers circular holes of very high quality that might have an interesting potential for high-precision applications.
Fig. 10. Microscope image of a matrix of drilled holes in fused silica of 75 µm thickness. The drilling parameters were 1000 bursts containing 50 pulses, burst fluence 219 J/cm2.
4. Conclusion
In conclusion, we demonstrated for the first time to our knowledge top-down percussion drilling of holes in dielectric materials in GHz-burst mode. Moreover, we studied the interaction process of the GHz-burst laser beam with the glass material sodalime by time-solved pump-probe shadowgraphy and thermal imaging revealing that the interaction process is fundamentally different for single-pulse and GHz-burst drilling. In GHz-burst mode, the drilling process is progressive whereas the matter is ejected in bunches. We confirm the thermal character of the interaction in GHz-burst mode leading to heat accumulation that has to be taken into account when choosing the drilling parameters. The obtained holes in GHz-burst drilling show a very high quality and aspect ratio and may pave the way for new applications where such extreme geometries are required.
Funding
Ministère de l'Économie, des Finances et de l'Industrie (172906051, 172906052, 172906053).
Acknowledgments
The authors thank Christophe Pradère and Alain Sommier from the I2M-TREFLE, University of Bordeaux, for the assistance in the thermal imaging experiments.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. C. Kerse, H. Kalaycoglu, P. Elahi, B. Cetin, D. Kesim, O. Akçaalan, S. Yavas, M. Asik, B. Oktem, H. Hoogland, R. Holzwarth, and F. O. Ilday, “Ablation-cooled material removal with ultrafast bursts of pulses,” Nature 537(7618), 84–88 (2016). [CrossRef]
2. P. Elahi, O. Akçaalan, C. Ertek, K. Eken, F. O. Ilday, and H. Kalaycoglu, “High-power Yb-doped all-fiber laser delivering 300 fs pulses for high-speed ablation-cooled materiel removal,” Opt. Lett. 43(3), 535–538 (2018). [CrossRef]
3. G. Bonamis, K. Mishchik, E. Audouard, C. Hönninger, E. Mottay, J. Lopez, and I. Manek-Hönninger, “High efficiency femtosecond laser ablation with gigahetz level burst,” J. Laser Appl. 31(2), 022205 (2019). [CrossRef]
4. K. Mishchik, G. Bonamis, J. Qiao, J. Lopez, E. Audouard, E. Mottay, C. Hönninger, and I. Manek-Hönninger, “High-efficiency femtosecond ablation of silicon with GHz repetition rate laser source,” Opt. Lett. 44(9), 2193–2196 (2019). [CrossRef]
5. G. Bonamis, E. Audouard, C. Hönninger, J. Lopez, K. Mishchik, E. Mottay, and I. Manek-Hönninger, “Systematic study of laser ablation with GHz bursts of femtosecond pulses,” Opt. Express 28(19), 27702–27714 (2020). [CrossRef]
6. T. Hirsiger, M. Gafner, S. M. Remund, M. W. Chaja, A. Urniezius, S. Butkus, and B. Neuenschwander, “Machining metals and silicon with GHz bursts: Surprising tremendous reduction of the specific removal rate for surface texturing applications,” Proc. SPIE 11267, 27 (2020). [CrossRef]
7. A. Žemaitis, M. Gaidys, P. Gečys, M. Barkauskas, and M. Gedvilas, “Femtosecond laser ablation by bibursts in the MHz and GHz pulse repetition rates,” Opt. Express 29(5), 7641–7653 (2021). [CrossRef]
8. D. J. Förster, B. Jaeggi, A. Michalowski, and B. Neuenschwander, “Review on Experimental and Theoretical Investigations of Ultra-Short Pulsed Laser Ablation of Metals with Burst Pulses,” Materials 14(12), 3331 (2021). [CrossRef]
9. S. M. Remund, M. Gafner, M. V. Chaja, A. Urniezius, S. Butkus, and B. Neuenschwander, “Milling applications with GHz burst: Investigations concerning the removal rate and machining quality,” Procedia CIRP 94, 850–855 (2020). [CrossRef]
10. S. Schwarz, S. Rung, C. Esen, and R. Hellmann, “Enhanced ablation efficiency using GHz bursts in micromachining fused silica,” Opt. Lett. 46(2), 282–285 (2021). [CrossRef]
11. S. Hendow, H. Takahashi, M. Yamaguchi, and J. Xu, “Enhanced ablation using GHz-pulsed fs laser,” Proc. SPIE 11268, 7 (2020). [CrossRef]
12. D. Esser, S. Rezaei, J. Li, P. R. Herman, and J. Gottmann, “Time dynamics of burst-train filamentation assisted femtosecond laser machining in glasses,” Opt. Express 19(25), 25632–25642 (2011). [CrossRef]
13. D. Grossmann, M. Reininghaus, C. Kalupka, M. Jenne, and M. Kumkar, “In-situ microscopy of front and rear side ablation processes in alkali aluminosilicate glass using ultra short pulsed laser radiation,” Opt. Express 25(23), 28478–28488 (2017). [CrossRef]
14. K. Sugioka, “Will GHz burst mode create a new path to femtosecond laser processing?” International Journal of Extreme Manufacturing, IJEM-110242 (2021).
15. D. Nieto, J. Arines, G. O’Connor, and M. T. Flores-Arias, “Single-pulse laser ablation threshold of borosilicate, fused silica, sapphire, and soda-lime glass for pulse widths of 500 fs, 10 ps, 20 ns,” Appl. Opt. 54(29), 8596–8601 (2015). [CrossRef]
16. F. Bauer, A. Michalowski, T. Kiedrowski, and S. Nolte, “Heat accumulation in ultra-short pulsed scanning laser ablation of metals,” Opt. Express 23(2), 1035–1043 (2015). [CrossRef]
