Open Access
Add to BibSonomyAbstract
We report on an Yb:YAG thin-disk multipass laser amplifier delivering sub-8 ps pulses at a wavelength of 1030 nm with 1420 W of average output power and 4.7 mJ of pulse energy. The amplifier is seeded by a regenerative amplifier delivering 6.5 ps pulses with 300 kHz of repetition rate and an average power of 115 W. The optical efficiency of the multipass amplifier was measured to be 48% and the beam quality factor was better than M2 = 1.4. Furthermore we report on the external second harmonic generation from 1030 nm to 515 nm using an LBO crystal leading to an output power of 820 W with 2.7 mJ of energy per pulse. This corresponds to a conversion efficiency of 70%. Additionally, 234 W of average power were obtained at the third harmonic with a wavelength of 343 nm.
© 2015 Optical Society of America
1. Introduction
In recent years material processing with ultra-short laser pulses has been a rapidly growing field. One very promising application for example is cutting and structuring of carbon fiber reinforced plastics (CFRP) [1]. Here, high average output powers lead to higher processing speeds while pulse energies in the mJ regime open possibilities for processing with larger spots and for multi-spot parallel processing. Furthermore, moderate repetition rates are very promising to decrease the thermal damaging due to heat accumulation caused by consecutive pulses in the processed material [2].
There are different approaches to developing laser systems for ultra-short pulses with very high average power. One is to directly scale the output power of mode-locked oscillators. An important prerequisite are lasers with very high cw output powers in fundamental-mode operation [3,4] and recently more than 4 kW of output power were demonstrated [5] using a single thin-disk laser. In mode-locked operation output powers of >270 W [6, 7] and pulse energies exceeding 80 µJ [8] were demonstrated with different setups. The pulse energy of these approaches is currently limited by the strong self-phase modulation (SPM) occurring at high intracavity intensities, and the repetition rate is fixed by the resonator length. In parallel, new materials like Yb:Lu2O3 or Yb:CALGO are investigated to further decrease pulse duration with thin-disk laser oscillators [9, 10].
Other concepts include amplifiers, such as fiber amplifiers where 640-fs pulses with an average output power of up to 830 W and a pulse energy of 10.6 µJ were reported [11]. Furthermore, using a coherent combination technique, 530 W of average power with 1.3 mJ of pulse energy was obtained with fiber amplifiers [12]. Reducing nonlinearities thanks to longer pulse durations, an output power of 2 kW at a pulse duration of 150 ps was demonstrated recently with a pulse energy of 100 µJ [13] and a beam quality factor M2<3.
Furthermore, using slab amplifiers, 1.1 kW of average output power, 55 µJ of pulse energy, and a pulse duration of 615 fs were demonstrated with M2x = 2.7 and M2y = 1.5 [14]. Additionally, the same group demonstrated 930 W of output power (800 fs, 46.5 µJ) with M2x = 1.42 and M2y = 1.09 [15]. Both concepts, slab and fiber amplifiers, show high average output powers but scaling the pulse energy to the mJ range requires chirped pulse amplification (CPA) or coherent combination.
Another interesting approach is to use amplifiers based on the thin-disk technology. Regenerative amplifiers for ultra-short pulses deliver average output powers exceeding 600 W and pulse energies exceeding 100 mJ [16]. However, without CPA we expect that the average output power for few-ps pulses achievable with this approach will be limited to approximately 1 kW by the currently available electro-optic modulators.
The thin-disk multipass amplifier avoids the use of such limiting elements as the number of passes on the disk – and therefore the gain – is scaled by angular multiplexing of the seed beam on the disk. This allows scaling of the output power to the multi-kW level with an almost independent scalability of the pulse energy to very high values. In addition to this operation at low repetition rates, the multipass amplifier approach shows no fundamental limit for high repetition rates. Due to their limited gain these devices need powerful seed sources in order to saturate the gain. This type of amplifier has been used for ns pulses [17] as well as to amplify 800 µs long bursts of sub-ps pulses [18, 19]. Using this thin-disk multipass amplifier approach we recently reported on an average output power of 1.1 kW with 1.4 mJ of pulse energy and a pulse duration of 7.3 ps [20].
In addition to infrared operation, high power frequency-doubled and -tripled ultrafast lasers are also very promising to exploit the higher absorption coefficients of some materials (e.g. copper and certain polymers) in the green and UV spectral range. The shorter wavelength further allows decreasing the focal diameter which is favourable to drill very small holes [21]. Intracavity second harmonic generation (SHG) was reported recently to produce up to 1.8 kW of output power at 515 nm with a pulse length of a few 100 ns and an optical efficiency of 30% with respect to the pump power at a wavelength of 938 nm [22]. Additionally, more than 500 W with less than 20 ns of pulse duration at a wavelength of 343 nm were demonstrated in the same work. Frequency-doubling of sup-ps pulses recently led to the demonstration of a green average power of 445 W at pulse repetition rate of 20 MHz [23].
In the UV spectral range, 39.1 W of output power and 39.1 µJ of pulse energy were demonstrated for 12 ps long infrared pulses [24]. The highest UV average power demonstrated so far with ultrashort pulses is 63 W at a repetition rate of 80 MHz (0.8 µJ) using a fiber master oscillator power amplifier with 40 ps of pulse duration [25]. In terms of UV pulse energy the record so far is 2 mJ converting 9.3 ps long infrared pulses with a repetition rate of 1 kHz which corresponds to 2 W of average output power [26].
In the following we report on a thin-disk multipass amplifier for sub 8-ps pulses with an average output power of 1420 W and at the same time 4.7 mJ of pulse energy at the fundamental wavelength of 1030 nm. These laser pulses were frequency doubled to produce 820 W of average power at the second-harmonic wavelength of 515 nm. In a further step, 234 W of output power at the third-harmonic wavelength of 343 nm were generated with a pulse energy of 780 µJ. To the best of our knowledge these results are new power records for sub-100 ps lasers system at all three wavelengths.
2. Experimental setup and results
2.1 The multipass amplifier
The seed laser generates pulses with a duration of 6.5 ps at an average output power of 115 W. In comparison to the previous investigations with 800 kHz [20] the repetition rate was now reduced to 300 kHz [27] to further explore the capability of energy scaling with the multipass thin-disk laser amplifier.
Figure 1 shows the setup of the multipass amplifier, which is identical to the one described in [20, 27]. The amplifier medium consists of an anti-reflection coated, plane-parallel Yb:YAG thin-disk (with a diameter of 17 mm, a radius of curvature of 20.5 m, a thickness of 130 µm, and 10-11% of doping concentration) mounted on a diamond heat sink by TRUMPF. The disk is pumped by a multi-pass cavity with 24 passes (standard “G1-module” as commonly provided by IFSW). The pumped spot has a diameter of approximately 6-7 mm with up to 2.7 kW pump power. The pump diodes were developed at Trumpf and are stabilized at a wavelength of 969 nm using Volume Bragg Gratings.
Fig. 1 Setup of the multipass amplifier. A modified version of the commercially available TruMicro 5050 laser was used as the seed laser. In combination with the end mirror and the quarter-wave plate in front of it, the thin-film polarizer (TFP) enables a double-pass through the amplifier set-up. The multipass amplifier itself consists of an array with 40 mirrors (details see text and Fig. 2), the thin-disk in a pump cavity, a plane folding mirror M1 and the retroreflecting mirror pair (RMP) next to the disk.
To fold the seed beam path over the disk, an array of 40 individually adjustable plane mirrors (HR-coated for 1030 nm) is used. The mirror holders are mounted on a water-cooled copper block in a distance of 1170 mm from the disk. At the side of the pump cavity two plane HR-coated folding mirrors M1 and M20-21 as well as a retroreflecting mirror pair (RMP) are used (see [20] for more details).
The collimated seed beam is linearly polarized and is incident with a diameter of 5.5 mm. This corresponds to 70-80% of the pump spot diameter in order to prevent higher-order modes [3] from amplification while enabling best extraction of the energy stored inside the pumped volume. During the experiments, we slightly modified the initial beam diameter using a telescope. The chosen beam diameter of 5.5 mm proved to be a good choice as smaller beam diameters led to less energy extraction and larger beam diameters did not show much higher extraction but would lead to tighter foci along the beam propagation in the amplifier.
The propagation in the amplifier follows the scheme depicted in Fig. 2. The beam passes mirror 1 on the array and reaches mirror 2 after a reflection on the disk. Then, the beam passes the RMP and reaches mirror 3. After this, the beam is reflected on the disk, mirror 4, mirror M1 and mirror 5. This scheme is continued until mirror 40 is reached (after mirror 20 one bounce on mirror M20-21 instead of M1 is needed to access the upper and lower line of mirrors on the mirror array). Then, the beam passes a λ/4-waveplate and a plane HR-coated end mirror which sends the beam back through the amplifier with the orthogonal linear polarization state compared to the incoming beam which doubles the number of passes through the disk. In total this setup allows for an overall number of 40 reflections on the disk.
Fig. 2 Backside view of the mirror array including a projection of the thin-disk, the mirrors M1 and M20-21 and the RMP to illustrate the propagation in the amplifier in analogy to the original concept described in [17]. The seed pulse passes the components as follows: 1 ➔ thin-disk ➔ 2 ➔ RMP ➔ 3 ➔ thin-disk ➔ 4 ➔ M1 ➔ 5 ➔ thin-disk ➔ 6 and so forth. M20-21 is needed to access the upper and lower line of mirrors.
Along the propagation length of approximately 170 m in the amplifier the beam stays almost collimated due to the use of plane mirrors and the large ROC of the disk. Without thermal lensing on the disk the beam diameter was calculated to oscillate between 5.5 mm and 1.5 mm [Fig. 3(a)]. With increasing strength of the thermal lens induced by the high pump power the variation of the beam diameter along the beam path through the amplifier is strongly reduced, as indicated by Fig. 3(b) assuming a strong convex thermal lens of fthermal = −15 m for illustration purposes. With the concave focal length of the unpumped disk being fdisk = 10.25 m this leads to a concave focal length of about 32 m for the pumped disk. As an alternative, mirror M1 can be chosen with a long convex ROC to achieve the same effect. The comparably constant and large beam diameter is very favorable in comparison to conventional 4f-propagation schemes [28, 29] as the pickup of B-integral of the amplified beam is minimized and spectral broadening due to self-phase modulation stays negligible up to high output energies. Assuming a beam quality factor of M2 = 1, a repetition rate of 300 kHz and a pulse duration of 8 ps, the calculated B-integral in the air amounts to 1.9 mrad/W. The disk contributes another 0.2 mrad/W. If – in the sense of a worst-case scenario – the full output power of 1420 W would propagate through the amplifier from the very beginning, the B-integral would amount to 3 rad. Due to amplifier dynamics, a slightly larger M2 and the positive effect of thermal lensing this B-integral is significantly reduced. In comparison, the B-integral in a 4f-scheme would be about 4 times as high as the worst-case for the approach we utilized.
Fig. 3 Beam diameter depicted over the whole propagation length inside the multipass amplifier. Red lines indicate the position of the thin-disk, green lines positions of mirror M1, and blue lines of the RMP. a) Beam diameter without the influence of a thermal lens on the disk and with M2 = 1. b) Influence of a thermal lens of fthermal = −15 m (for illustration). The oscillation is less steep and the pick-up of B-integral is significantly reduced.
Additionally, in combination with thin-disk technology this multipass scheme easily allows further scaling of pulse energy by increasing the beam diameter and adapting the disk curvature or the ROC of mirror M1 accordingly. This avoids nonlinear limitations. Thus the need to use a vacuum environment for future energy scaling can be mitigated.
Due to the long propagation length, air flows have a large influence on the beam stability. In our set-up external air turbulences are shielded by an air-tight acrylic glass cover around the complete amplifier. Internal air flows, especially in front of the heated disk, are minimized by zero-phonon-line pumping [4, 20, 30, 31] and effectively compensated for by the RMP in the setup [20].
As shown in Fig. 4 we have demonstrated an output power of up to 1420 W with a repetition rate of 300 kHz. This corresponds to a pulse energy of 4.7 mJ. Assuming a Gaussian-shaped temporal pulse profile with a duration of 8 ps the peak power is calculated to be 0.56 GW which is much higher than the peak powers of comparable kW-level femtosecond fiber [11] or slab [14] amplifiers. The average power and the pulse energy were limited only by the maximum available pump power of 2700 W. With the seed power of 115 W being subtracted, the corresponding optical efficiency for the extracted power is 48%.
Fig. 4 Output powers of the amplifier with 40 passes over the disk for different seed powers and repetition rates. A maximum output power of 1420 W was reached with a seed power of 115 W at a pump power of 2700 W and 300 kHz repetition rate. The comparison between different repetition rates at the same seed power of 80 W shows the independent scalability of pulse energy and average output power in this regime.
The data points with an average seed power of 80 W but with different repetition rates of 800 kHz (1.38 mJ per pulse at 1.1 kW average power) [20] and 300 kHz (3.7 mJ per pulse at 1.1 kW average power) are shown in Fig. 4 to demonstrate the energy scalability of the approach. The fact that the efficiencies and maximum output powers are almost identical in the two cases illustrates that the multipass amplifier can independently scale the average output power and the pulse energy at least at repetition rates within the range of 300-800 kHz. Furthermore, we observed that the maximum amplification factor (ratio between maximum amplified output power and seed power) stays almost constant for different seed powers at a given number of passes over the disk. Thus, with 115 W of seed power it was possible to amplify the beam up to a power of 1.4 kW, limited only by available pump power. Experiments are under way to further increase the pump power and the number of amplifying passes of the laser beam on the disk.
A time-dependent measurement of the output power starting at 1070 W shows a slight drop by 7% to 1000 W after about eight minutes. This is accompanied by a shift of the beams on the disk. After switching off the pump diodes the beams return into their initial positions after some minutes. Therefore, we assume that this power drop is due to a thermal drift of some elements in the setup which can be mitigated by improved cooling. This will also be subject of future investigations.
The pulse duration was measured with a commercial APE Pulse Check. Assuming a Gaussian temporal profile, the duration of the seed pulses after propagation through the un-pumped amplifier was measured to be τ = 6.5 ps FWHM. As the seed laser was operated with energies significantly above the standard product specifications noticeable SPM-induced spectral broadening was observed. Due to the long-term power decrease of the amplifier the pulse duration, spectrum and beam quality were recorded slightly below maximum output power. At an amplified output power of 1.3 kW (due to the long-term power decrease of the amplifier) the pulse duration was measured to be τ = 7.7 ps (see Fig. 5) accompanied by only a slight spectral gain narrowing as can be seen in Fig. 6(a). The beam quality factor M2 was measured to be below 1.4 in both axes using a commercial WinCamD beam profiler according to ISO 11146. Figure 6(b) shows the beam caustic and the far field intensity distribution of the amplified beam, respectively. At high output powers we observe an onset of (most probably thermal) aberrations in the beam profile measured at the exit of the amplifier [see inset of Fig. 6(b)]. The resulting deviation from the Gaussian intensity profile therefore leads to local differences between fitted envelope and the actual beam diameter near the waist position. The beam diameter at the beam waist is actually smaller than the fitted envelope while the measured divergence angle fits quite well. The M2-value of 1.4 can therefore be considered to be an estimate for the upper limit, also because all measurements even at 1.4 kW always yielded an M2<1.4.
Fig. 5 Autocorrelation trace measured at 1.3 kW of output power. The pulse duration was measured to be τ = 7.7 ps.
Fig. 6 a) Spectra of the pulses after the amplifier for the seed without pumping and at 1340 W of average output power; b) beam quality measurement and collimated beam profile at the exit of the amplifier at 1.3 kW of output power.
2.2 Second harmonic generation
An LBO crystal (type I, critical phase matching) was used to frequency double the output beam of the amplifier as depicted in Fig. 7. The linearly polarized output beam leaving the amplifier passes a λ/2-waveplate in a motorized rotary stage enabling to direct the power either to a power meter (power head IR 1) or to the SHG setup. This setup allows aligning the components and the angular phase-matching of the crystal at low powers while the amplifier runs at full output power. This avoids any changes to the input beam that may occur when increasing its power to the maximum available value. Furthermore, it allows measuring the incident power at the fundamental wavelength. The second λ/2-waveplate in the SHG setup is used to ensure phase-matching in the LBO crystal. The beam diameter in the crystal is reduced from 4.5 mm to 2.25 mm by the telescope formed by two HR mirrors separated by 250 mm. The first mirror has a concave ROC of R = 1000 mm and the second one a convex ROC of R = −500 mm. A water-cooled aperture with a diameter of 5 mm is located in front of the LBO crystal to prevent any damage in case of unexpected beam drift or misalignment. The LBO itself has an aperture of 6 mm x 6 mm and a length of 5 mm with cut angles θ = 90° and Φ = 12.9° for type I phase-matching. It is AR-coated for the wavelengths of 1030 nm and 515 nm on both facets. The crystal was actively stabilized to a temperature of 47° C. After the frequency conversion process in the LBO the two wavelengths are separated by two dichroic mirrors at an angle of incidence of 45°.
Fig. 7 Setup used for SHG. The beam path can be switched by a λ/2-waveplate in a rotary stage and a TFP to either the power head or the SHG setup. The beam size at the LBO crystal is adjusted by a telescope. The output is analysed after a dichroic mirror to separate the two wavelengths.
The output powers were measured behind filter 1 and on both ports of filter 2. Furthermore, a beam quality measurement was performed for the green beam.
The measured output power at the second-harmonic wavelength of 515 nm versus the input power at the wavelength of 1030 nm (measured at the power head IR 1) is shown in Fig. 8. At a maximum fundamental input power of 1170 W the output power at 515 nm was measured to be 820 W, corresponding to 2.7 mJ of energy per pulse [27]. To the best of our knowledge this is the highest average output power reported for an ultrafast laser at this wavelength and corresponds to a conversion efficiency of 70%. The slight deviations from a linear curve are most probably due to the fact that the orientation of the crystal was not realigned with the full output power being directed to the SHG setup. At maximum green output power the residual IR power was measured at power head IR 3 to be 2.5 W leading to the conclusion that the infrared power content in the measurement depicted in Fig. 8 is negligibly small.
Fig. 8 Output power at the wavelength of 515 nm (diamonds) and conversion efficiency (rectangles) versus input power at the fundamental wavelength of 1030 nm measured at the power head next to the TFP. At 1170 W of input power the output power was measured to be 820 W which corresponds to a conversion efficiency of 70%.
The beam caustic was measured at the maximum green average power of 820 W by means of the WinCamD beam profiler with a band pass filter for 515 nm. The beam profiles measured at a distance of about 16 mm behind the beam waist position are depicted in Fig. 9(a) for 500 W and in Fig. 9(b) for 820 W of green output power. At 500 W only small thermal aberrations are observed. The beam quality deteriorates at higher output powers. This effect appears to be stronger on one axis. At 820 W of green output power we measured M2x = 1.53 and M2y = 1.99. With misalignments of the optical path, we measured slightly higher M2-values. This suggests that one reason for the aberrations can be a slight misalignment or shift of the infrared beam which leads to diffraction at the aperture. Additionally, we believe that is caused by thermally induced phase mismatches in the LBO crystal [23] or conversion saturation effects also contribute to the deterioration in beam quality. This will be subject to future investigations. LBO crystals with larger apertures will be employed as they allow larger beam diameters in the crystal. With this the thermal load is distributed over a larger area and conversion saturation is shifted towards higher IR input powers and thus to higher green output powers.
Fig. 9 Beam profile about 16 mm behind the beam waist position for a) 500 W and b) 820 W of green output power. A deterioration in beam quality can be observed.
2.3 Third harmonic generation
The scheme for the third-harmonic generation (THG) (sum-frequency generation between 1030 nm and 515 nm) is depicted in Fig. 10. The main difference to the setup for SHG (Fig. 7) is the second LBO-crystal placed behind the first LBO crystal for SHG. This second crystal has an aperture of 6 mm x 4 mm (width x height) and a length of 10 mm. The crystal is cut with angles θ = 51.7 and Φ = 90° for type II phase-matching (oeo) of the sum-frequency generation between the beams at 1030 nm and 515 nm to a wavelength of 343 nm. Both facets are AR-coated for all three wavelengths. The crystal is placed inside an oven and temperature-stabilized to 47° C.
Fig. 10 Setup used for THG. The main difference to Fig. 7 is a second temperature-stabilized LBO crystal which was added. Using this crystal a sum-frequency generation (SFG) between the 1030 nm and 515 nm was performed to create UV-light at 343 nm. Additionally, dichroic mirrors to separate the UV wavelength and a beam stabilization device were implemented. Furthermore, the telescope was changed to adapt the beam diameter inside the crystals.
The telescope was changed to a combination of a concave (ROC = 750 mm) and a convex (ROC = −250 mm) mirror in a distance of 250 mm to each other. Thus, the collimated beam diameter of the incident fundamental wave is now reduced from 4.5 mm to 1.5 mm and is adapted to the aperture of the available crystal. Additionally, this allows for increased conversion efficiency, especially at lower input powers, as the setup was initially optimized to provide a UV output power of 100 W with reasonable conversion efficiency.
Furthermore, two additional high-power suitable dichroic mirrors were used to separate the UV beam from the green and IR beam. The UV beam was then measured using an additional power head. To verify the functionality of the dichroic mirrors, the second LBO was removed and the residual power at the UV power head was measured to be 0.7 W at 775 W of IR input power. This proves that the IR and green content in the UV power measurement is negligibly small.
For safety reasons a light-tight box was mounted around the setup for protection against UV stray light. A dichroic mirror (HR for UV, AR for IR, 0° angle of incidence) was placed at the entrance hole of the box to reflect any UV light back inside the box.
The setup was further optimized in comparison to the SHG setup by using a commercial “Aligna” beam stabilization system behind the amplifier output to stabilize pointing and position of the IR beam at the entrance facets of the crystals.
Figure 11(a) depicts the output power at 343 nm and the UV conversion efficiency versus the input power at 1030 nm. An output power of 234 W was obtained at 725 W of input power. This is - to the best of our knowledge - the highest UV output power reported for an ultrashort laser system so far. Taking into account the repetition rate of 300 kHz this corresponds to a pulse energy of 780 µJ. The conversion efficiency, which was extracted without taking into account the losses of the IR beam caused by the optics in front of the crystals, was 32% at maximum output power.
Fig. 11 a) Output power (diamonds) and conversion efficiency (rectangles) at the wavelength of 343 nm versus the input power at 1030 nm. A maximum output power of 234 W was achieved with 32% conversion efficiency. b) Output power (diamonds) and conversion efficiency (rectangles) at the wavelength of 515 nm with an input beam size of 1.5 mm and without the second LBO crystal.
As can be seen from Fig. 11(a) the highest efficiency of 41% was obtained at 483 W of IR input power. The decreased conversion efficiency at higher powers was investigated by two additional experiments.
First, the second LBO was removed to measure the SHG power generated with the smaller 1.5 mm beam diameter used for the THG setup. This revealed that the reduced beam diameter leads to a reduced SHG efficiency starting at IR power levels beyond about 600 W [see Fig. 11(b)]. This is most probably attributed to thermally induced phase-mismatches or conversion saturation as discussed in the SHG experiments before. In future setups this can be mitigated by using crystals with larger apertures and larger beam sizes.
The second test was devoted to the analysis of the intensity distribution of the UV beam by a camera which was placed behind an HR-mirror MCAM shown in Fig. 10. The distance between the second LBO crystal and the camera is about 1 m. As can be seen in Fig. 12(a), no obvious signs of beam quality degradation occur up to about 120 W of generated UV output power. At higher powers aberrations as shown in Fig. 12(b) can be observed. Apart from influences of a degrading beam quality of the green beam this may be attributed to saturation effects, a power ratio mismatch between the IR and the green beam, absorption induced stress in the crystal or a non-optimal temperature regulation of the oven. The crystal is sandwiched in the oven between two Peltier-elements and is actively controlled to 47° C using a PT100 temperature measurement element placed between the crystal and the Peltier-elements. At high powers the temperature had however to be reduced to about 30° C to achieve the highest conversion efficiency. Hence one may assume that the type II LBO crystal is subject to excessive heating caused by absorption of the high powers leading to an inhomogeneous spatial temperature profile inside the crystal causing thermally induced optical aberrations.
Fig. 12 a) Image of the UV beam (observed at a distance of about 1 m from the second LBO crystal behind a HR-mirror steering the beam to the UV power head) at 120 W of output power. No beam degradation was observed. b) Far-field image at 234 W of output power. The beam profile shows strong signs of aberrations.
To test the long-term stability of the THG setup, the UV output power generated with an IR input power of 300 W was monitored for a duration of 25 minutes. As shown in Fig. 13 the UV power undergoes a slow variation which however correlates with the variation of the IR input simultaneously measured at the power head IR 1 (being a qualitative signal only). The small power drop at the end could be compensated for by switching off the power and readjusting the temperature by about 1 K without changing the spot on the crystal. This leads to the conclusion that the THG crystal does not suffer any degradation (e.g. due to color centers) within the 25 minutes of the experiment.
Fig. 13 UV output power measurement over 25 minutes. Additionally, the infrared input power was monitored using power head IR 1 behind the TFP. The UV output power follows the input power over time and no power drop due to crystal degradation was observed on this time-scale.
4. Conclusions and outlook
In conclusion, we demonstrated an ultrafast thin-disk multipass laser amplifier delivering 1.4 kW of average output power with pulses with 4.7 mJ of energy and a duration of 8 ps (at a repetition rate of 300 kHz). The beam quality factor was better than M2 = 1.4. The experiments show that the thin-disk multipass amplifier can scale pulse energy and average output power independently, at least for the investigated repetition rates between 300 and 800 kHz.
Frequency doubling by means of an LBO crystal led to 820 W of average power at a wavelength of 515 nm with 1170 W of incident IR power which corresponds to a conversion efficiency of 70% and a SHG pulse energy of 2.7 mJ.
By sum-frequency generation between the beams at 1030 nm and 515 nm in a second LBO crystal, an average UV power of 234 W (780 µJ of pulse energy) was generated at a wavelength of 343 nm with a conversion efficiency of 32%.
To the best of our knowledge these are the highest average output powers reported for ultrafast laser systems with pulses shorter than 100 ps so far for all three wavelengths. The infrared output power is currently only limited by the available pump power for the multipass amplifier while the output powers in the green and UV spectral region are limited by thermal effects and the apertures of the crystals employed.
Future work will focus on using shorter seed pulses as well as increasing the output power by implementing a higher number of passes in the amplifier and the pump module and by increasing the pump power. For the higher-harmonic generation crystals with larger apertures and an improved temperature control will be implemented to further improve the performance.
Acknowledgments
This work was funded by the German Federal Ministry of Education and Research (BMBF) under contract number 13N11787 and 13N11789. Furthermore, we would like to thank Michael Glatt for his helpful work in the construction of the setup and in the experiments.
This work was supported by the German Research Foundation (DFG) within the funding program Open Access Publishing.
References and links
1. C. Freitag, M. Wiedenmann, J. Negel, A. Loescher, V. Onuseit, R. Weber, M. Abdou Ahmed, and T. Graf, “High quality processing of CFRP with a 1.1 kW picosecond laser,” Appl. Phys., A Mater. Sci. Process. 119(4), 1237–1243 (2015). [CrossRef]
2. 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 (May 2014) +, Erratum Optics Express 22(23), 28232–28233 (2014). [CrossRef]
3. S. Piehler, B. Weichelt, A. Voss, M. A. Ahmed, and T. Graf, “Power scaling of fundamental-mode thin-disk lasers using intracavity deformable mirrors,” Opt. Lett. 37(24), 5033–5035 (2012). [CrossRef] [PubMed]
4. B. Weichelt, A. Voss, M. Abdou Ahmed, and T. Graf, “Enhanced performance of thin-disk lasers by pumping into the zero-phonon line,” Opt. Lett. 37(15), 3045–3047 (2012). [CrossRef] [PubMed]
5. S. Schad, V. Kuhn, T. Gottwald, V. Negoita, A. Killi, and K. Wallmeroth, “Near fundamental mode high-power thin-disk laser,” Proc. SPIE 8959, Solid State Lasers XXIII: Technology and Devices, 89590U (February 28, 2014).
6. C. J. Saraceno, F. Emaury, O. H. Heckl, C. R. Baer, M. Hoffmann, C. Schriber, M. Golling, T. Südmeyer, and U. Keller, “275 W average output power from a femtosecond thin disk oscillator operated in a vacuum environment,” Opt. Express 20(21), 23535–23541 (2012). [CrossRef] [PubMed]
7. J. Brons, V. Pervak, E. Fedulova, D. Bauer, D. Sutter, V. Kalashnikov, A. Apolonskiy, O. Pronin, and F. Krausz, “Energy scaling of Kerr-lens mode-locked thin-disk oscillators,” Opt. Lett. 39(22), 6442–6445 (2014). [CrossRef] [PubMed]
8. C. J. Saraceno, F. Emaury, C. Schriber, M. Hoffmann, M. Golling, T. Südmeyer, and U. Keller, “Ultrafast thin-disk laser with 80 μJ pulse energy and 242 W of average power,” Opt. Lett. 39(1), 9–12 (2014). [CrossRef] [PubMed]
9. S. Ricaud, A. Jaffres, K. Wentsch, A. Suganuma, B. Viana, P. Loiseau, B. Weichelt, M. Abdou-Ahmed, A. Voss, T. Graf, D. Rytz, C. Hönninger, E. Mottay, P. Georges, and F. Druon, “Femtosecond Yb:CaGdAlO4 thin-disk oscillator,” Opt. Lett. 37(19), 3984–3986 (2012). [CrossRef] [PubMed]
10. C. R. Baer, O. H. Heckl, C. J. Saraceno, C. Schriber, C. Kränkel, T. Südmeyer, and U. Keller, “Frontiers in passively mode-locked high-power thin disk laser oscillators,” Opt. Express 20(7), 7054–7065 (2012). [CrossRef] [PubMed]
11. T. Eidam, S. Hanf, E. Seise, T. V. Andersen, T. Gabler, C. Wirth, T. Schreiber, J. Limpert, and A. Tünnermann, “Femtosecond fiber CPA system emitting 830 W average output power,” Opt. Lett. 35(2), 94–96 (2010). [CrossRef] [PubMed]
12. A. Klenke, S. Breitkopf, M. Kienel, T. Gottschall, T. Eidam, S. Hädrich, J. Rothhardt, J. Limpert, and A. Tünnermann, “530 W, 1.3 mJ, four-channel coherently combined femtosecond fiber chirped-pulse amplification system,” Opt. Lett. 38(13), 2283–2285 (2013). [CrossRef] [PubMed]
13. H. J. Otto, F. Stutzki, N. Modsching, C. Jauregui, J. Limpert, and A. Tünnermann, “2 kW average power from a pulsed Yb-doped rod-type fiber amplifier,” Opt. Lett. 39(22), 6446–6449 (2014). [CrossRef] [PubMed]
14. P. Russbueldt, T. Mans, J. Weitenberg, H. D. Hoffmann, and R. Poprawe, “Compact diode-pumped 1.1 kW Yb:YAG Innoslab femtosecond amplifier,” Opt. Lett. 35(24), 4169–4171 (2010). [CrossRef] [PubMed]
15. P. Russbueldt, D. Hoffmann, M. Hofer, J. Lohring, 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]
16. T. Metzger, M. Gorjan, M. Ueffing, C. Teisset, M. Schultze, R. Bessing, M. Häfner, S. Prinz, D. Sutter, K. Michel, H. Barros, Z. Major, and F. Krausz, “Picosecond Thin-Disk Lasers,“ Conference on Lasers and Electro-Optics CLEO 2014, invited paper JTh4L.
17. A. Antognini, K. Schuhmann, F. D. Amaro, F. Biraben, A. Dax, A. Giesen, T. Graf, T. W. Hänsch, P. Indelicato, L. Julien, C.-Y. Kao, P. E. Knowles, F. Kottmann, E. Le Bigot, Y.-W. Liu, L. Ludhova, N. Moschüring, F. Mulhauser, T. Nebel, F. Nez, P. Rabinowitz, C. Schwob, D. Taqqu, and R. Pohl, “Thin-Disk Yb:YAG Oscillator-Amplifier Laser, ASE, and Effective Yb:YAG Lifetime,” IEEE J. Quantum Electron. 45(8), 993–1005 (2009). [CrossRef]
18. M. Schulz, R. Riedel, A. Willner, S. Düsterer, M. J. Prandolini, J. Feldhaus, B. Faatz, J. Rossbach, M. Drescher, and F. Tavella, “Pulsed operation of a high average power Yb:YAG thin-disk multipass amplifier,” Opt. Express 20(5), 5038–5043 (2012). [CrossRef] [PubMed]
19. M. Schulz, H. Hoeppner, M. Temme, R. Riedel, B. Faatz, M. J. Prandolini, M. Drescher, and F. Tavella, “14 kilowatt burst average power from 2-stage cascaded Yb:YAG thin-disk multipass amplifier,” in Frontiers in Optics 2013, I. Kang, D. Reitze, N. Alic, and D. Hagan, eds., OSA Technical Digest (online) (Optical Society of America, 2013), paper FTu4A.2.
20. J. P. 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] [PubMed]
21. A. Feuer, C. Kunz, M. Kraus, V. Onuseit, R. Weber, T. Graf, D. Ingildeev, and F. Hermanutz, “Influence of laser parameters on quality of microholes and process efficiency,” Proc. 8967, 89670H (2014).
22. C. Stolzenburg, W. Schüle, V. Angrick, M. Bouzid and A. Killi, “Multi-kW IR and green nanosecond thin-disk lasers,” Proc. SPIE 8959, Solid State Lasers XXIII: Technology and Devices, 89590O (February 28, 2014).
23. B. Gronloh, P. Russbueldt, B. Jungbluth, and H. Hoffmann, “Green sub-ps laser exceeding 400 W of average power,” Proc. SPIE 8959, Solid State Lasers XXIII: Technology and Devices, 89590T (February 28, 2014).
24. P. Zhu, D. Li, Q. Liu, J. Chen, S. Fu, P. Shi, K. Du, and P. Loosen, “39.1 μJ picosecond ultraviolet pulses at 355 nm with 1 MHz repeat rate,” Opt. Lett. 38(22), 4716–4718 (2013). [CrossRef] [PubMed]
25. D. Sangla, J. Saby, B. Cocquelin, and F. Salin,”High power picosecond fiber laser emitting 50 W at 343 nm at 80 MHz,” Proc. SPIE 8237, Fiber Lasers IX: Technology, Systems, and Applications, 82370N (February 9, 2012).
26. D. Chen, Z. Bai, Y. Pan, M. Chen, and G. Li, “Two-millijoule, 1 kHz, 355-nm picosecond laser pulse generation in LiB3O5 crystal,” Opt. Eng. 52(8), 086107 (2013). [CrossRef]
27. J. Negel, A. Loescher, A. Voss, M. Abdou Ahmed, D. Bauer, D. Sutter, A. Killi, and T. Graf, “Ultrafast thin-disk amplifier with 1.4 kW average power and 4.7 mJ pulse energy at 1030 nm converted to 820 W and 2.7 mJ at 515 nm,” Proc. SPIE 9342, Solid State Lasers XXIV: Technology and Devices, 93420V (26 March 2015).
28. D. Müller, S. Erhard, O. Ronsin, and A. Giesen, “Thin disk multi-pass amplifier,” OSA Trends in Optics and Photonics. 1, paper 278 (2003).
29. J. Neuhaus, D. Bauer, J. Zhang, A. Killi, J. Kleinbauer, M. Kumkar, S. Weiler, M. Guina, D. H. Sutter, and T. Dekorsy, “Subpicosecond thin-disk laser oscillator with pulse energies of up to 25.9 microjoules by use of an active multipass geometry,” Opt. Express 16(25), 20530–20539 (2008). [CrossRef] [PubMed]
30. J. J. Romero, J. Johannsen, M. Mond, K. Petermann, G. Huber, and E. Heumann, “Continuous-wave laser action of Yb3+-doped lanthanum scandium borate,” Appl. Phys. B 80(2), 159–163 (2005). [CrossRef]
31. D. Sangla, M. Castaing, F. Balembois, and P. Georges, “Highly efficient Nd:YVO4 laser by direct in-band diode pumping at 914 nm,” Opt. Lett. 34(14), 2159–2161 (2009). [CrossRef] [PubMed]
