Open Access
Add to BibSonomy
Abstract
A highly productive ablation process of 100 nm thick platinum films with a processed area rate of up to 378 cm2/min is presented using radially and azimuthally polarized laser beams. This was achieved by developing a laser amplifier generating 757 fs long laser pulses at a maximum average power of 390 W and a repetition rate of 10.6 MHz with adjustable polarization states, i.e., linear, radial, and azimuthal polarization on the work piece. The pulse train emitted from the laser was synchronized to a custom-designed polygon scanner and directed into an application machine.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
During the last decades, ultrafast laser sources gained considerable attention thanks to their versatile applications in industrial micro material processing such as structuring of surfaces [1], drilling of holes with high aspect ratio [2] or cutting of carbon fiber reinforced plastics (CFRP) [3]. One important economic factor for an application process is the processing time. Since, according to obvious energetic considerations, the processing time can be decreased by increasing the average laser power, one driving force in the field of laser development is the establishment of lasers with high average power. To further optimize the application process and maintain an optimum removal rate, the pulse repetition rate has to be adapted to the average laser power [4]. Thus, the repetition rate or the diameter of the beam on the work piece has to be increased to maintain the optimum fluence when raising the average laser power. However, since the desired feature size of the structures produced on the work piece defines the usable beam diameter [5], the repetition rate and the scanning speed have to be adapted accordingly, in order to prevent undesired heat accumulation and degradation of the quality of the processed parts [6]. Once the optimum process parameters are found, the realization of high throughput is essential to enter into the field of industrial applications such as e.g. in the chip production, where high productivity is essential to reduce the overall price per unit. Additionally, laser micro-machining is a green solution to enhance the productivity in selective thin-film removal compared to standard processes such as photolithography.
Several systems have demonstrated the capability to deliver ultrashort pulses with average output powers at a kilowatt level [7–12]. All these systems are based on the master oscillator power amplifier (MOPA) architecture and delivered linearly polarized pulses. Changing the linear polarization state into a radial or azimuthal polarization state and transforming the Gaussian intensity profile into a doughnut-shaped intensity distribution [13] have attracted great interest in the field of material processing [14–20]. The influence of the polarization state on the process is highlighted for example by drilling results in steel [18], structuring of Ni-surfaces [19], and structuring of Silicon-surfaces [20]. Moreover, the edges of the intensity or fluence distributions are steeper compared to the ones of Gaussian beams. Hence, it is possible to create finer structures and sharper edges in laser material processing due to the sharper edges of the fluence distribution. The highest average power for ultrashort radially polarized pulses directly generated by a laser oscillator so far is 125 W at a repetition rate of 78 MHz and a pulse duration of 970 fs [21]. The corresponding pulse energy is 1.6 µJ. Another approach to generate powerful radially or azimuthally polarized ultrashort pulses is the polarization conversion of powerful linearly polarized ultra-short pulses [13,18,22,23]. In a subsequent step, these pulses may be amplified further. Different approaches and systems have been reported in the literature. The highest peak power demonstrated so far is 90 GW at a repetition rate of 3 kHz and with an average power of 2.4 W generated by a Ti:sapphire laser system [24]. A step-index few-mode fibre laser delivered 106 W of average output power at a repetition rate of 5.5 MHz and a pulse duration of 110 ps [25]. An average power of 85 W was demonstrated at a repetition rate of 20 MHz and with sub-1 ps pulse duration using an Yb:YAG single-crystal fiber amplifier [26]. Various experiments were performed with the thin-disk multi-pass architecture. An Yb:YAG thin-disk multi-pass amplifier (TD-MPA) system was reported to generate 235 W of average output power at a repetition rate of 200 kHz and with sub-1 ps pulse duration [27]. A similar system demonstrated 265 W of average output power at a repetition rate of 20 MHz and a pulse duration of 782 fs [28]. In addition, a similar system generated 416 W of average output power with sub-ps and radially polarized pulses [29]. Radially polarized pulses with a pulse duration of less than 10 ps were extracted from a TD-MPA at an average power of 635 W and with a repetition rate of 300 kHz [30].
In the following we present the productive laser ablation of platinum structures on silica substrates by 757 fs long pulses with a repetition rate of 10.6 MHz and an average output power of up to 390 W from a TD-MPA. By implementation of a polarization-based multiplexing approach for the ultra-short seed pulses, a pulse-on-demand scheme operating at 10.6 MHz was achieved. This facilitated the use of a high-speed polygon scanner operating at a scan speed of 100 m/s, which was synchronized to the laser system.
2. Implementation of the laser system and machine
2.1 Master oscillator power amplifier system
The laser system, which was implemented for our material processing demonstration, is a master oscillator power amplifier (MOPA) composed of a seed laser consisting of a bulk SESAM mode-locked laser oscillator and a single-crystal fiber (SCF) amplifier [26] followed by a thin-disk multi-pass power booster as the final amplification stage. The laser’s setup is illustrated in Fig. 1. The seed laser system (laser oscillator plus SCF) runs at a central wavelength of 1029.5 nm and generates pulses with an average power of 48 W with a pulse duration of 675 fs at a repetition rate of 10.6 MHz. The pulse picking needed for the materials processing was implemented by a polarization-multiplexing scheme that ensures a constant saturation of the TD-MPA that avoids the risk to damage the components of the amplifier which may occur when the seed beam is interrupted while the amplifier is still pumped (high population inversion without extraction). This is realized by launching both of the beams deflected into the 0th and the 1st diffraction of the acousto-optical modulator (AOM) shown in Fig. 1 into the TD-MPA. By rotating the polarization of one of the two beams (0th or 1st order) one can combine those by using a thin-film polarizer in order to always seed the thin-disk amplifier independently from the deflection at the AOM. Since the AOM is placed at the early stage in the amplifier chain where the power and energy are moderate, it can be operated at a very high modulation speed (> multi-MHz, ≈ pulse on demand). Since the diffraction efficiency of the pulse picker (AOM) is less than 100%, it is the beam from the 1st diffraction order that was directed towards the application stage after the TD-MPA, in order to ensure a high contrast of the laser pulses on the work piece. The corresponding maximum average power in the 1st diffraction order is 38 W, whereat approx. 10 W of average power remains in the 0th order transmitted beam. The beam that is launched into the TD-MPA is first shaped and collimated to a diameter of 3.8 mm by a set of corresponding mirrors. The TD-MPA itself is based on the concept already presented in [12]. Due to the comparatively large beam diameters inside the amplifier, chirped-pulse amplification (CPA) is not required. An Yb:YAG thin-disk laser crystal, which is HR-coated on its back side and AR-coated on the front side and with a radius of curvature (RoC) of 22 m was used. It has a thickness of 170 µm and a doping concentration of approx. 11 at. %. For efficient heat removal and high stiffness, the disk was glued on a diamond heat sink. The thin-disk laser crystal was implemented into a multi-pass pumping cavity allowing for 24 pump passes. To minimize the heat load inside the Yb:YAG disk, the pumping occurs at a wavelength of 969 nm (“zero phonon line”) by using a spectrally stabilized laser diode provided by DILAS. The diameter of the pump spot was set to be 5.2 mm on the thin-disk laser crystal. The seed beam is folded over the thin-disk crystal 30 times by using an array of 60 adjustable plane mirrors mounted on a water-cooled copper block. This mirror array is located at a distance of 1150 mm from the disk while the additional folding mirrors (M1-3) as well as the retro-reflecting mirror pair (RMP) are positioned at a distance of 950 mm from the mirror array. This leads to a total beam propagation length of approx. 130 m inside the TD-MPA.
Fig. 1. Setup of the laser system. The pulses emitted from the mode-locked seed laser (oscillator plus SCF amplifier) are launched into the thin-disk multi-pass amplifier (TD-MPA). The latter is composed of an array with 60 mirrors, a thin-disk laser crystal inside the pump module, plane folding mirrors M1-3, the retro-reflecting mirror pair (RMP), two telescopes consisting of the mirrors with radii of curvature (RoC) of -2 m and 3 m, thin-film polarizers (TFP), and half-wave plates (HWP). The latter allow to adjust the power directed to the material processing station. A linear to radial/azimuthal polarization converter (LRAC) is placed into the beam path to the application station.
After 30 reflections of the beam on the disk’s back side, a thin-film polarizer (TFP2) reflects the vertically polarized (0th diffraction order of the AOM) pulses towards a power head. The horizontally polarized pulses (1st diffraction order of the AOM) are directed towards the materials processing station. The average amplified power can be adjusted by the combination of a HWP and TFP3. The beam reflected from TFP 3 was also used for diagnostics. A linear to radial or azimuthal polarization converter (LRAC) is placed in the beam path to the application station. This LRAC is composed of 8 half-wave plate segments similar to the one used in [18,23,30]. The advantage of this approach is the flexibility to obtain linear, radial or azimuthal polarization states on the work piece.
The amplification results are shown in Fig. 2. The output power is plotted versus the incident pump power on the thin disk. The black squares represent the output power and amplification efficiency obtained with the vertically (s-) polarized (0th order) pulses with an average seed power of 48 W. An average output power of 445 W at 895 W of incident pump power could be extracted corresponding to an optical efficiency of 44.4%, with the seed power being subtracted. The red triangles show the output power and amplification efficiency of the horizontally (p-) polarized (1st order) picked pulses having an average power of 38 W (diffraction efficiency of 79% = 38/48). In this case an average output power of 390 W at an incident pump power of 895 W was reached corresponding to an optical efficiency of 39.2%, with the seed power being subtracted. At the same time the remaining power of 10 W ( = 48 W - 38 W) still leaked into the 0th diffraction order of the AOM which was amplified to 48 W by the TD-MPA.
Fig. 2. Output power versus incident pump power of the TD-MPA with 30 reflections on the thin-disk laser crystal for different inputs. A maximum output power of 445 W was reached with an average seed power of 48 W. 390 W of average power were achieved for the horizontally polarized “picked” pulses with a seed power of 38 W. The right axis shows the amplification efficiencies with the corresponding seed power being subtracted.
Figure 3 shows the caustic of the output beam that was later used for materials processing, recorded at an average output power of 390 W along with the intensity distribution at three different positions. The beam diameter was measured by using a commercial WinCamD beam profiler on a translation stage and the beam quality resulted to be M2x = 1.15 in the horizontal axis and M2y = 1.3 in the vertical axis. The fact that the output beam was astigmatic can be explained by a non-perfectly stigmatic seed beam as shown with the caustic around the waist position presented in Fig. 4.
Fig. 3. a) Caustic of the output beam at 390 W of average power. Intensity distribution at different positions along the caustic b) at 0.8 mm, c) at 7.4 mm, and d) at 15.6 mm.
Fig. 4. a) Caustic of the seed beam after the AOM (1st order) at 38 W of average power around focal positions of x- and y-axis. Intensity distribution at different positions along the caustic b) at 4.3 mm, c) at 7.3 mm, and d) at 10.3 mm.
The pulse duration was measured by means of a commercial APE pulseCheck autocorrelator. Assuming a sech2-temproal profile, the duration of the seed pulses after propagation through the unpumped amplifier was measured to be 676 fs full-width at half maximum (FWHM). At 390 W of average output power the pulse duration increased to 757 fs (see Fig. 5(a)). This slight increase by 81 fs arises from spectral gain narrowing when pumping the thin-disk laser crystal. Figure 5(b)) depicts the recorded spectrum at 390 W of output power. The central wavelength is 1029.5 nm and the spectral FWHM is 1.5 nm. This would correspond to a transform-limited pulse duration of 743 fs, which is consistent with the pulse duration retrieved from the autocorrelation trace measurement.
Fig. 5. Autocorrelation trace a) and spectrum b) of the output pulses measured at 390 W of average power. A sech2-fit reveals a pulse duration of 757 fs. The central wavelength is 1029.5 nm and the FWHM is 1.5 nm.
The amplifier’s long-term power stability was also tested since a high power stability is important in particular in order to obtain reliable processing results. The test was performed at a pump power of the TD-MPA of 800 W over a duration of 235 min. Figure 6 depicts the recorded output power (blue points) over a time interval of 135 min with the warm-up time of 100 min not included.
Fig. 6. Average output power as a function of time. The power was measured during 235 min, while the warm-up time of 100 min is not included in the plot. The average power value of the horizontally polarized output beam is 324.5 W with a standard deviation of 1.4 W (0.43%).
The average value of the power during this time interval was 324.5 W and the standard deviation was 1.4 W, which corresponds to a relative standard deviation of 0.43%. The peak-to-valley value was found to be 8.9 W, which corresponds to a relative deviation of 2.7%. Operating the laser at pump powers exceeding 800 W led to a slight decrease of power over time. This can be attributed to thermal instabilities of some of the mechanical parts composing the TD-MPA.
2.2 Polarization shaping of the amplified beam
In order to investigate the influence of radially or azimuthally polarized ultrashort pulses on the laser ablation of platinum layers, a linear to radial or azimuthal polarization converter (LRAC) was placed into the beam path to the application station. The LRAC transforms the linearly polarized Gaussian fundamental mode (TEM00) into a cylindrically polarized Laguerre-Gaussian mode LG01*. The beam diameter on the polarization converter was 5.5 mm. A free-space propagation distance of 5 m was implemented behind the LRAC in order to clean-up the beam. In addition, a water-cooled aperture with a diameter of 20 mm, placed at a distance of 400 mm behind the LRAC, was used to filter the light scattered from the converter. The beam profile measured at a power of 310 W is shown in Fig. 8(a). In this case, the main axis of the polarization converter was adjusted in such a way to obtain a beam with azimuthal polarization. This latter was analyzed qualitatively with a linear polarizer on a rotary stage and quantitatively with a 2D-Polarimeter [31]. The degree of azimuthal polarization (after the beam clean-up) was measured to be 98% [32]. The M2 was measured to be Mx2=2.5 and My2=2.4 (it is worth to mention that the beam clean-up system was not perfect which mainly explains the higher M2 in comparison to the theoretical M2 value of 2). The available power with radial or azimuthal polarization as a function of the pump power of the TD-MPA is presented in Fig. 7. As a reference, the power was measured without the polarization converter i.e. for linear polarization (black squares). The polarization conversion efficiency is shown on the right axis in Fig. 7. At maximum power a conversion efficiency of around 95% was obtained. The mean value is 94.9% when all data points are taken into account. The error bar exceeds 1% for values below 70 W. This can be explained by a measured power fluctuation of 0.5 W for this measurement. By controlling the orientation of the incident linear polarization on the converter either a radially or an azimuthally polarized beam can be generated. Figure 8 shows the far field beam profiles of the azimuthally (a) and radially (c) polarized beams at an average power of 310 W. The panels b) and d) of Fig. 8 show qualitative polarization analyses which were recorded by rotating a linear polarizer in front of the beam profiler. Panels e) and f) present the corresponding simulation results for our polarization conversion scheme. The electrical field and the intensity distribution are calculated using the field tracing algorithm provided in VirtualLab [33]. The simulation is performed as follows: A fundamental mode with a beam roundness of 80% (close to the beam obtained experimentally out of the front-end amplifier) propagates through a segmented wave plate (8-segments) and the resulting intensity distribution is observed after 5 m of propagation in free space (Fig. 8(e)). In order to analyze the polarization purity, a linear polarizer is inserted in the beam path in front of the detector (Fig. 8(f)). As one can see, the calculated intensity distributions are in good agreement with the experiments.
Fig. 7. Power of the beams after conversion to either radial or azimuthal polarization and conversion efficiency.
Fig. 8. Far-field beam profiles of the azimuthally (a) and radially (c) polarized beam at 310 W of power. A qualitative polarization analysis is shown in b) for the azimuthally polarized beam and in d) for the radially polarized beam. The panels e) and f) show corresponding simulations. The white arrows indicate the orientation of the linear polarizer.
3. Demonstration of an application
For application in materials processing, the laser system described above was integrated into a laser machine (GL.compact, GFH GmbH, Deggendorf, Germany). This machine was also equipped with a polygon scanner (LSE 170 HNA, Next Scan Technology, Evergem, Belgium), which was synchronized with the pulse picker of the laser. The axis of the laser machine, which moved the polygon scanner perpendicular to its scan direction, was also synchronized with the scanner system. The samples to be processed were 6” silica wafers, which were coated with a 100 nm thick platinum layer. The microstructured samples can be used as a circuit board, for instance in the field of Lab-on-a-Chip devices. For these devices, high quality and a high precision of the microstructured conductor paths are essential. To compare with conventional manufacturing methods, a high process speed is required as well, to ensure low production costs. Using conventional ultrafast laser systems, the process speed is principally limited by the available average laser power, but also by the repetition rate and the reachable scanning speeds of conventional galvanometer scanners. The effective processed area rate for the microstructuring of the platinum-coated wafers was limited to 21.6 cm2/min when we used an industrial benchmark laser system (TruMicro 5050 Femto Edition, Trumpf GmbH & Co. KG, Ditzingen, Germany) with a repetition rate of 800 kHz, linear polarization, Gaussian intensity distribution (M2 ≈ 1.1), a pulse duration of 900 fs, an average power of 40 W, and using a galvanometer scanner for beam deflection (intelliSCAN, SCANLAB GmbH, Puchheim, Germany).
The process parameters for the ablation process of platinum using this benchmark system are listed in Table 1. These parameters were obtained after a process development in several trials, to ensure an effective ablation of the platinum.
Table 1. Benchmark process parameters for patterning silica-wafers with a 100 nm platinum coating
Figure 9 shows the patterning quality using the benchmark process parameters on the platinum-coated silica wafers. Our experiments showed that the quality at the edge of the remaining platinum is reduced when applying low pulse overlap to attain higher processed area rates. The quality can be improved, for instance by a multi-pass process or by reducing the scanning speed at the same repetition rate. However, this would result in lower processed area rates.
Fig. 9. Ablation results of the platinum-coated silica wafer using the benchmark laser system (EP = 5 µJ).
To ensure a stable process using the TD-MPA, the laser system was operated at constant average output power of 320 W, which corresponds to a pulse energy (EP) of 30.5 µJ at the repetition rate of 10.6 MHz. The average power on the work piece was adjusted by the HWP in front of TFP 3 (see Fig. 1). The beam transmitted through the polygon scanning unit was measured to attain a maximum average power of 205 W which reveals a significant loss of 34% introduced by the components used in the polygon scanner.
The focus position with respect to the work piece was determined before processing the coated substrates by moving the scanning unit vertically in steps of 100 µm while pattering single lines on the sample, see Fig. 10. In this case, the sample was an aluminum coated glass wafer with a 100 nm thick metallic layer. The laser beam was radially polarized with a pulse energy of 6.8 µJ. With the focus on the surface of the work piece the width of the ablated lines was measured to be about 32 µm.
Fig. 10. Single lines patterned on an Al coated (100 nm) glass wafer, with varying focal position (radially polarized, EP = 6.8 µJ).
The optimum process parameters for the ablation process of the platinum film using the TD-MPA are listened in Table 2. These parameters were obtained after a process development in several trials.
Table 2. Optimized process parameters for patterning silica-wafers with a 100 nm platinum coating
The optimum pulse energy for the process was determined by varying the pulse energy from 1.6 µJ to 9.5 µJ. Single lines with a pulse overlap of about 72% (diameter in scan direction) were written. This was done for both radially and azimuthally polarized beams and for a linearly polarized Gaussian beam. In the latter case, the polarization converter was not inserted in the beam path. It was found that a minimum pulse energy of 2.3 µJ was required for processing the 100 nm thick platinum film in our experiments. To improve the patterning quality in our experiments, the pulse energy was increased to 4.2 µJ for the microstructuring of the platinum coated wafers. While on the one hand a further increase of the pulse energy on the sample led to damages on the substrate’s surface, on the other hand defocusing decreased the patterning quality. Figure 11 shows three processed areas obtained when using the linearly polarized Gaussian beam (a), the radially polarized doughnut beam (b), and the azimuthally polarized doughnut beam (c) at a constant pulse energy of 4.2 µJ. The resulting width of the ablated line was 22.8 µm, 31.9 µm, and 32.5 µm, respectively.
Fig. 11. Single line ablation on a platinum coated wafer at constant pulse energy of 4.2 µJ: a) for linearly, b) radially, and c) for azimuthally polarized light.
The optimized process parameters given in Table 2 were used for patterning Lab-on-a-chip (LOC) wafers. Figure 12 shows the patterning result of a LOC-glass wafer coated with 100 nm of platinum using the radial polarization. The magnified picture c) shows that the conductor paths on the platinum wafer show a high patterning quality without visible damages on the glass carrier. The effective processed area rate for this process was 378 cm2/min. Thus, the processing time for such wafers was reduced by a factor of more than 17 compared to the benchmark process and at the same time exhibits a higher patterning quality, especially at the edge of the remaining platinum layer. This higher quality is achieved thanks to the intensity profile of the radially/azimuthally polarized beams, which exhibit steeper flanks and thus, sharper edges with no residues inside the ablated path. By comparison, the edge quality of the path, which was structured by using linear polarization, shows platinum residues, which extend up to 4 µm into the ablated path. However, the beam diameter on the work piece is increased when using beams with radial/azimuthal polarizations (LG01*) instead of linear polarization (Gaussian) due to the different focusing properties of these laser beams.
Fig. 12. Resistance sensor on a platinum Lab-on-a-Chip wafer patterned with the radially polarized laser beam. a) Overview of the bitmap design, b) & c) single paths of the sensor.
4. Conclusion
In conclusion, we demonstrated a highly productive laser ablation process of 100 nm thin platinum films on silica wafers using beams with radial/azimuthal polarization. We reached an effective processed area rate of 378 cm2/min at higher patterning qualities, which also enables a more than 17 times faster microstructuring process, compared to the benchmark laser system used for this specific process.
Different approaches of thin-film ablation were investigated in Ref. [34] and in Ref. [35]. Both references introduce strategies to achieve high quality thin-film ablation without damaging the substrate underneath. The influence of process parameters using a Gaussian intensity distribution is explained in details for ablation of 100 nm NiCr-films in Ref. [34]. The authors in [35] introduced an approach to lift-off up to 100 nm thin gold layers by taking advantage of strong electron-phonon coupling in titanium. Here, the titanium was used as transition layer to the glass substrate. In contrast to our results, the laser sources in [34] and [35] delivered maximal pulse repetition rates of 0.6 MHz and 1 MHz, respectively, and the scanner achieved scanning speeds of approx. 1-3 m/s.
Our promising results were achieved thanks to the implementation of a thin-disk multi-pass amplifier delivering radially/azimuthally polarized pulses at an average power of 390 W, a repetition rate of 10.6 MHz, and a pulse duration of 757 fs. To obtain the pulse on demand scheme, these pulses were modulated at high-speed using a robust polarization multiplexing scheme of the seed pulses. This approach presents a solution to current limitations in state-of-the-art laser micromachining at multi-MHz repetition rate as reported in [36], for example. The main limitation of our demonstrated process is the scanning speed of the polygon scanner. Higher scanning speeds would increase the processed area rate, since our laser system can deliver sufficient energy per pulse for this specific process.
Future work will focus on further power scaling and exploring the benefit of radially/azimuthally polarized laser beam of ultra-short pulses for other applications such as drilling of high aspect ratio hole or texturing of large area for superhydrophoby application. Implementation of additional flexibility of the laser system in terms of pulse shaping shall also be possible with our approach.
Funding
Seventh Framework Programme (619237, Ultrafast Razipol project).
Disclosures
The authors declare no conflicts of interest.
References
1. S. Nolte, C. Momma, H. Jacobs, A. Tünnermann, B. N. Chichkov, B. Wellegehausen, and H. Welling, “Ablation of metals by ultrashort laser pulses,” J. Opt. Soc. Am. B 14(10), 2716–2722 (1997). [CrossRef]
2. D. Breitling, C. Föhl, F. Dausinger, T. Kononenko, and V. Konov, in Femtosecond Technology for Technical and Medical Applications, F. Dausinger, F. Lichtner, and H. Lubatschowski, eds. Springer, (2004).
3. R. Weber, C. Freitag, T. V. Kononenko, M. Hafner, V. Onuseit, P. Berger, and T. Graf, “Short-pulse laser processing of CFRP,” Phys. Procedia 39, 137–146 (2012). [CrossRef]
4. B. Neuenschwander, B. Jaeggi, M. Schmid, and G. Hennig, “Surface structuring with ultra-short laser pulses: Basics, limitations and needs for high throughput,” Phys. Procedia 56, 1047–1058 (2014). [CrossRef]
5. B. Jaeggi, B. Neuenschwander, U. Hunziker, J. Zuercher, T. Meier, M. Zimmermann, K. H. Selbmann, and G. Hennig, “Ultra-high-precision surfacestructuring by synchronizing a galvo scanner with an ultra-short-pulsed lasersystem in MOPA arrangement,” Proc. SPIE 8243, 82430K (2012). [CrossRef]
6. R. Weber, T. Graf, P. Berger, V. Onuseit, M. Wiedenmann, C. Freitag, and A. Feuer, “Heat accumulation during pulsed laser materials processing,” Opt. Express 22(9), 11312–11324 (2014). [CrossRef]
7. M. Müller, A. Klenke, A. Steinkopff, H. Stark, A. Tünnermann, and J. Limpert, “3.5 kW coherently combined ultrafast fiber laser,” Opt. Lett. 43(24), 6037–6040 (2018). [CrossRef]
8. J. Negel, A. Loescher, D. Bauer, D. Sutter, A. Killi, M. A. Ahmed, and T. Graf, “Second Generation Thin-Disk Multi-pass Amplifier Delivering Picosecond Pulses with 2 kW of Average Output Power,” in Lasers Congress 2016 (ASSL, LSC, LAC), OSA Technical Digest (online) (Optical Society of America, 2016), paper ATu4A.5 (2016).
9. M. Müller, M. Kienel, A. Klenke, T. Gottschall, E. Shestaev, M. Plötner, J. Limpert, and A. Tünnermann, “1 kW 1 mJ eight-channel ultrafast fiber laser,” Opt. Lett. 41(15), 3439–3442 (2016). [CrossRef]
10. P. Russbueldt, D. Hoffmann, M. Höfer, J. Löhring, J. Luttmann, A. Meissner, J. Weitenberg, M. Traub, T. Sartorius, D. Esser, R. Wester, P. Loosen, and R. Poprawe, “Innoslab Amplifiers,” IEEE J. Sel. Top. Quantum Electron. 21(1), 447–463 (2015). [CrossRef]
11. T. Nubbemeyer, M. Kaumanns, M. Ueffing, M. Gorjan, A. Alismail, H. Fattahi, J. Brons, O. Pronin, H. G. Barros, Z. Major, T. Metzger, D. Sutter, and F. Krausz, “1 kW, 200 mJ picosecond thin-disk laser system,” Opt. Lett. 42(7), 1381–1384 (2017). [CrossRef]
12. J. Negel, A. Voss, M. Abdou Ahmed, D. Bauer, D. Sutter, A. Killi, and T. Graf, “1.1 kW average output power from a thin-disk multipass amplifier for ultrashort laser pulses,” Opt. Lett. 38(24), 5442–5445 (2013). [CrossRef]
13. Q. Zhan, “Cylindrical vector beams: from mathematical concepts to applications,” Adv. Opt. Photonics 1(1), 1–57 (2009). [CrossRef]
14. J. Lopez, G. Mincuzzi, R. Devillard, Y. Zaouter, C. Hönninger, E. Mottay, and R. Kling, “Ablation efficiency of high average power ultrafast laser,” J. Laser Appl. 27(S2), S28008 (2015). [CrossRef]
15. R. Weber, A. Michalowski, M. Abdou Ahmed, V. Onuseit, V. Rominger, M. Kraus, and T. Graf, “Effects of Radial and Tangential Polarization in Laser Material Processing,” Phys. Procedia 12(A), 21–30 (2011). [CrossRef]
16. J. J. J. Nivas, F. Cardano, Z. Song, A. Rubano, R. Fittipaldi, A. Vecchione, D. Paparo, L. Marrucci, R. Bruzzese, and S. Amoruso, “Surface Structuring with Polarization-Singular Femtosecond Laser Beams Generated by a q-plate,” Sci. Rep. 7(1), 42142 (2017). [CrossRef]
17. G. Niziev and A. V. Nesterov, “Influence of beam polarization on lasercutting efficiency,” J. Phys. D32, 1455–1461 (1999). [CrossRef]
18. M. Kraus, M. Abdou Ahmed, A. Michalowski, A. Voss, R. Weber, and T. Graf, “Microdrilling in steel using ultrashort pulsed laser beams with radial and azimuthal polarization,” Opt. Express 18(21), 22305–22313 (2010). [CrossRef]
19. E. Skoulas, A. Manousaki, C. Fotakis, and E. Stratakis, “Biomimetic surface structuring using cylindrical vector femtosecond laser beams,” Sci. Rep. 7(1), 45114 (2017). [CrossRef]
20. J. J. J. Nivas, H. Shutong, K. K. Anoop, A. Rubano, R. Fittipaldi, A. Vecchione, D. Paparo, L. Marrucci, R. Bruzzese, and S. Amoruso, “Laser ablation of silicon induced by a femtosecond optical vortex beam,” Opt. Lett. 40(20), 4611–4614 (2015). [CrossRef]
21. F. Beirow, M. Eckerle, B. Dannecker, T. Dietrich, M. Abdou Ahmed, and T. Graf, “Radially polarized passively mode-locked thin-disk laser oscillator emitting sub-picosecond pulses with an average output power exceeding the 100 W level,” Opt. Express 26(4), 4401–4410 (2018). [CrossRef]
22. M. Beresna, M. Gecevičius, P. G. Kazansky, and T. Gertus, “Radially polarized optical vortex converter created by femtosecond laser nanostructuring of glass,” Appl. Phys. Lett. 98(20), 201101 (2011). [CrossRef]
23. G. Machavariani, Y. Lumer, I. Moshe, A. Meir, and S. Jackel, “Efficient extracavity generation of radially and azimuthally polarized beams,” Opt. Lett. 32(11), 1468–1470 (2007). [CrossRef]
24. S. Carbajo, E. A. Nanni, L. J. Wong, G. Moriena, P. D. Keathley, G. Laurent, R. J. D. Miller, and F. X. Kärtner, “Direct longitudinal laser acceleration of electrons in free space,” Phys. Rev. Accel. Beams 19(2), 021303 (2016). [CrossRef]
25. D. Lin, N. Baktash, S. Alam, and D. J. Richardson, “106 W, picosecond Yb-doped fiber MOPA system with a radially polarized output beam,” Opt. Lett. 43(20), 4957–4960 (2018). [CrossRef]
26. F. Lesparre, J. T. Gomes, X. Délen, I. Martial, J. Didierjean, W. Pallmann, B. Resan, M. Eckerle, T. Graf, M. Abdou Ahmed, F. Druon, F. Balembois, and P. Georges, “High-power Yb:YAG single-crystal fiber amplifiers for femtosecond lasers in cylindrical polarization,” Opt. Lett. 40(11), 2517–2520 (2015). [CrossRef]
27. J.-P. Negel, A. Loescher, B. Dannecker, P. Oldorf, S. Reichel, R. Peters, M. Abdou Ahmed, and T. Graf, “Thin-disk multi-pass amplifier for fs pulses delivering 400 W of average and 2.0 GW of peak power for linear polarization as well as 235 W and 1.2 GW for radial polarization,” Appl. Phys. B 123(5), 156 (2017). [CrossRef]
28. A. Loescher, J. Negel, T. Graf, W. Pallmann, B. Resan, I. Martial, J. Didierjean, F. Lesparre, J. Gomes, X. Delen, F. P. Druon, F. Balembois, P. Georges, and M. Abdou Ahmed, “A 265 W and 782 fs amplified radially polarized beam emitted by a thin-disk multipass amplifier,” ASSL, OSA Technical Digest (online) (Optical Society of America, 2015), paper ATh3A.3 (2015).
29. M. Abdou Ahmed, final press release of Ultrafast Razipol project, http://www.razipol.eu/public/D8.14_-Final_press_release.pdf (2017).
30. A. Loescher, J. Negel, T. Graf, and M. Abdou Ahmed, “Radially polarized emission with 635 W of average power and 2.1 mJ of pulse energy generated by an ultrafast thin-disk multi-pass amplifier,” Opt. Lett. 40(24), 5758–5761 (2015). [CrossRef]
31. T. Liebig, M. Abdou Ahmed, A. Voss, and T. Graf, “Novel multi-sensor polarimeter for the characterization of inhomogeneously polarized laser beams,” in SPIE LASE, Photonics West (2010).
32. A. Voss, M. Abdou Ahmed, and T. Graf, “Application of the extended Jones matrix formalism for higher-order transverse modes to laser resonators,” Opt. Express 18(21), 21540–21550 (2010). [CrossRef]
33. LightTrans GmbH, LightTrans VirtualLab Advanced, www.lighttrans.com, accessed (July, 2020).
34. C. von der Heide, M. Grein, G. Bräuer, and A. Dietzel, “Methodology of selective metallic thin film ablation from susceptible polymer substrate using pulsed femtosecond laser,” Opt. Express 28(22), 33413–33432 (2020). [CrossRef]
35. B. Kim, H. K. Nam, S. Watanabe, S. Park, Y. Kim, Y.-J. Kim, K. Fushinobu, and S.-W. Kim, “Selective Laser Ablation of Metal Thin Films Using Ultrashort Pulses,” Int. J. of Precis. Eng. and Manuf.-Green Tech., 2198, (2020).
36. B. Jaeggi, S. Remund, R. Streubel, B. Goekce, S. Barcikowski, and B. Neuenschwander, “Laser micromachining of metals with ultra-short pulses: factors limiting the scale-up process,” J. Laser Micro/Nanoeng. 12(3), 267–273 (2017).
