We present the results from a new frontend within a double-chirped pulse amplification architecture (DCPA) utilizing crossed-polarized wave generation (XPW) for generating ultra-high contrast, 150 μJ-level, femtosecond seed pulses at 1030 nm. These pulses are used in the high energy class diode-pumped laser system Polaris at the Helmholtz Institute in Jena. Within this frontend, laser pulses from a 75 MHz oscillator-pulse train are extracted at a repetition rate of 1 Hz, temporally stretched, amplified and then recompressed reaching a pulse energy of 2 mJ, a bandwidth of 12 nm and 112 fs pulse duration at a center wavelength of 1030 nm. These pulses are temporally filtered via XPW in a holographic-cut BaF2 crystal, resulting in 150 μJ pulse energy with an efficiency of 13 %. Due to this non-linear filtering, the relative intensity of the amplified spontaneous emission preceding the main pulse is suppressed to 2×10−13. This is, to the best of our knowledge, the lowest value achieved in a high peak power laser system operating at 1030 nm center wavelength.
© 2014 Optical Society of America
Ever since the technique of chirped pulse amplification was first introduced into laser systems , the attainable laser pulse energies have increased constantly. Nowadays state-of-the-art high-power laser systems can easily generate peak powers of several 100 TW or even 1 PW. When such laser pulses are focused onto a target (e.g. thin foils or gas jets) intensities exceeding 1021 W/cm2 can be reached [2, 3]. Under these extreme focussing conditions, the coherent optical radiation preceding the main pulse, such as amplified spontaneous emission (ASE) or pre-pulses, can be intense enough to influence or even suppress the interaction process under investigation, even though the associated intensity of such prepulses is lower by many orders of magnitude. Since these effects like pre-plasma formation or target heating have to be avoided in order to perform sophisticated experiments, the so called temporal intensity contrast (TIC) defined as the ratio of the main pulse intensity and of the parasitic pre- and post-pulses is an important parameter for such a laser system.
During the last decade, many schemes aiming at the improvement of the TIC, such as saturable absorbers (SA) , Kerr-gating , picosecond-pump OPA [6, 7] or frequency doubling , were proposed and tested. Another very prominent method for contrast improvement is the nonlinear filtering via cross-polarized wave generation (XPW)  in a so-called double chirped pulse amplification (DCPA) setup . Here, a second CPA stage is added to the laser system, in which the oscillator pulse is first temporally stretched, then amplified to the mJ-level and then recompressed to the femtosecond range, completing the first CPA stage. Thereafter, the nonlinear filter is applied to clean the pulse and thus increasing the TIC. After this cleaning the pulse is then sent to the second CPA stage, where the pulse is further stretched, amplified up to the Joule level, and re-compressed.
It has already been shown that setups utilizing XPW as the nonlinear filter are capable of improving the intensity contrast by several orders of magnitude [11, 12]. Additionally, this leads to a shortening of the compressed pulse duration due to spectral broadening occurring during the filtering process . Besides birefringence (e.g. induced by stress), nonlinear polarization rotation  in air or ionization effects, the contrast enhancement achievable by XPW is mainly limited by the extinction ratio of the crossed polarizers which are used to separate the input signal and the filtered signal. Therefore, state-of-the art polarizers and optics have to be used. To avoid any non-linearities caused in air the crystal has to be placed in a vacuum environment.
In the near infrared regime XPW has already been used to produce 37 μJ, 115 fs (at a center wavelength of 1030 nm)  and 180 μJ, 180 fs (1057 nm)  high-contrast pulses. However, for seeding high-intensity diode-pumped, solid-state (DPSS) laser systems in the 1 μm range, laser pulses with energies exceeding 100 μJ are necessary which have a sufficiently broad bandwidth to reach a pulse duration for the fully amplified pulse below 150 fs.
In this paper we present the application of the double-CPA scheme including a nonlinear filtering stage using XPW in BaF2 to the fully diode-pumped solid state laser POLARIS [2, 17], operated at the Institute of Optics and Quantum Electronics and at the Helmholtz-Institute Jena, Germany. The first CPA stage comprising an Öffner-type stretcher, a regenerative amplifier and a grating compressor is described and characterized in detail. The performance of the XPW filter was designed to match the input requirements of the second Polaris amplifier (A2) in terms of pulse energy and beam shape. In other words, the filtering stage had to deliver pulses of up to 150 μJ pulse energy and a homogeneous Gaussian shaped transverse profile at a repetition rate of 1 Hz in long term operation. We demonstrate an enhancement of the TIC by more than 4 orders of magnitude for the output pulses of Polaris, when seeded with the new XPW frontend. This was achieved by precisely adjusting the polarizers and using special polarization maintaining optics within and after the first compressor resulting in a polarization extinction ratio better than 5×10−6. The ASE pedestal of the amplified pulse has been reduced to an unprecedented relative intensity level of 2×10−13.
2. Experimental setup
2.1. Picosecond CPA stage
A sketch of the setup of the first CPA stage is shown in Fig. 1. This stage consists of a pulse-picker, an Öffner-type stretcher, which stretches the 100 fs oscillator pulses to about 20 ps, a regenerative amplifier and a grating compressor. The stretching factor has been chosen to be as small as possible, while allowing the pulse to be amplified to an energy of ≥2 mJ and the possibility to use small-diameter optics.
The femtosecond pulses are first picked off the 75 MHz train of a Ti:Sapphire Oscillator (Mira, Coherent Inc.) at a repetition rate of 1 Hz, before they are stretched to 20 ps in a 2-pass Öffner-type stretcher. The stretcher comprises a 50 × 50 mm2 gold coated grating (G1) with a groove density of 1200 lines/mm, an Öffner-telescope (OT) made out of a 50.8 mm round concave mirror with a focal length of 200 mm and a 25 × 10 mm2 square convex mirror (f=−100 mm), and a silver coated roof mirror (RM1) providing the second pass. The overall footprint of this first stretcher is as small as 40 × 20 cm2.
As the pulse is temporally stretched when entering the first amplifier, it can now be amplified up to 3 mJ, without damaging the optics or triggering nonlinear effects such as self-phase modulation. The beam diameter after the stretcher is adapted to the amplifier mode by a telescope (T1). A Faraday isolator (FI) protects the stretcher against back reflections from the amplifier, which is a regenerative ring cavity (setup details see Fig. 1). After 33 roundtrips the pulse is coupled out reaching a pulse energy of 3 mJ with an rms-energy stability of 3%. The roundtrip time of the amplifier can be matched to the oscillator repetition rate (time between two pulses: 13.6 ns), to avoid the generation of pre- and postpulses . The beam profile of the amplified pulse, displayed in Fig. 2(a), exhibits a Gaussian shape with a FWHM diameter of 1 mm. The spectrum of the amplified pulse shows a bandwidth (FWHM) of 12 nm (cf. Fig 2(b)), which by applying a Fourier-transform and assuming a sech2 pulse, is sufficient to support pulses as short as 98 fs after recompression (cf. Fig. 3). The spectral modulations are caused by interference between the main pulse and post-pulses in the picosecond range generated in transmitting optics. These postpulses are subsequently filtered by the XPW stage (cf. section 3.2). After amplification the beam is enlarged by another telescope (T2) and passes the first Glan-Laser polarizer (GLP). This GLP is the first polarizer of the XPW filter.
The compressor consists of two parallel gold coated gratings (G2,G3) (1200 lines/mm) separated by 95 mm and a roof mirror (RM2) for the second pass, reaching an energy throughput of 70 %. Behind the compressor the pulse duration can either be measured by a commercial single-shot autocorrelator (TOPAG Lasertechnik GmbH) or the pulses can be guided to the XPW-stage. The measured autocorrelation trace (see Fig. 3) shows a pulse duration of 112 fs, which is longer by 14 fs than the expected Fourier-limit, which is likely to be explained by uncompensated spectral phase.
2.2. Nonlinear filtering stage - the polarization extinction contrast
When using XPW filtering the achievable contrast enhancement is mainly limited by the extinction ratio of the two crossed polarizers separating the filtered signal from the input pulse. Therefore one has to use polarizers with extinction ratios between the two polarization states of 10−4 ... 10−5 or better.
The polarizers used here were BBO Glan-Laser polarizers with an extinction ratio of 5×10−6. This extinction ratio was measured including all optics (lenses, vacuum windows, crystal) using ns-pulses from the amplifier. To reduce the intensity experienced by the first polarizer, which might lead to unwanted spectral distortions or even damages of the uncoated surfaces inside the polarizer cube, this polarizer was placed in front of the compressor (see GLP Fig. 1). Since the laser pulses have a duration of about 20 ps at this point in the beamline, the intensity in the polarizer is reduced by a factor of ≈150 as compared to after compression. When using the XPW filter at intensities above 1012 W/cm2 for a few hours, dark spots appeared on both surfaces of the BaF2 crystal, which was also reported by Ramirez et al. . In Figure 4 two stages of the darkening are shown, where the crystal was used 4 hours (green circle) and more than 8 hours (red circle) with a 1 Hz pulse train. Reducing the vacuum pressure by 2 orders of magnitude and cleaning the tubes and chamber did not lead to an improvement. The darkening could only be avoided by increasing the spot size on the crystal and therefore reducing the intensity to ≈ 8.8 × 1011 W/cm2. At this intensity level no darkening could be observed after more than 40,000 shots.
As the first polarizer was placed in front of the compressor, some extra precautions had to be taken to maintain the high extinction ratio when the pulses pass through the compressor and the optics of the XPW stage. First, we used so-called zero-phase shift mirrors (ZPM) as turning mirrors, where the reflected beam does not experience a phase shift between s- and p-polarization. The roof mirror used in the compressor (RM2 Fig. 1) is a home made one out of ZPM’s. Another step was to match the polarization state of the GLP to that one supported by the compressor. This was done by turning the first GLP stepwise and minimizing the signal measured after the second polarizer at the end of the XPW stage. Extra care had to be taken when mounting the vacuum windows. A nonuniform tightening of the screws holding the window flange may lead to stress induced birefringence which was directly measurable by a decrease in the polarization contrast. This can easily lead to a reduction of the extinction ratio by more than one order of magnitude.
Behind the BaF2 crystal a thin film polarizer (TFP) is positioned to filter the XPW signal from the pulse (see Fig. 5(a)). Thin film polarizers are usually not capable of generating a polarization extinction better than 10−3, when used in pairs. However, if the incident light is already (linearly) polarized with a certain purity, in our case better than 10−5, one can use a single TFP as the analyzing polarizer, when it is precisely aligned and the XPW-signal is used in transmission. The TFP has a front side coating which is highly reflective for s-polarization at an angle of incidence of γ = 65°. This angle is rather uncritical, as the s-polarization remains parallel to the plane of the TFP surface, when γ changes. For achieving a high polarization contrast, the alignment of the angle ϑ, which is a rotation perpendicular to the plain of incidence, has to be realized with an accuracy of < 1mrad, as shown in Fig. 5.
When ϑ changes, the polarization (of the signal-pulse) incident on the TFP has an s- and a p-component, thus resulting in a lower polarization contrast. With the methods described above, a polarization contrast below 5×10−6 of the complete XPW setup including the compressor was achieved.
2.3. Nonlinear filtering stage - XPW generation
The setup of the XPW stage is shown in Figure 6. After passing through a 1m focussing lens (L1), the pulses enter the vacuum chamber through a fused silica window (FSW). The XPW is generated in a 2 mm thick -BaF2 crystal. The vacuum windows are wedged to avoid the generation of postpulses, especially by the XPW pulse. Another lens (L2) recollimates the beam and a following TFP separates the two polarizations.
Figure 7(a) shows the output beam profile of the XPW signal. Behind L2 it is a smooth Gaussian beam with an FWHM of 1 mm. In Fig. 7(b) the efficiency of the XPW process is shown. The energy of the XPW starts to saturate above 70 μJ, reaching more than 200 μJ at 18 % conversion efficiency, where the losses of the crystal and TFP are not considered. Energy efficiencies of up to 28% have been shown by Julien et al.  and were addressed to four-wave mixing processes such as self-phase modulation. However, at output energies above 150 μJ the beam profile starts to deform and the conversion process becomes unstable. This may be explained by nonlinear focusing and the onset of continuum generation. Therefore, the maximum usable energy is 150 μJ at a conversion efficiency of 13 %.
The spectrum of the pulse is broadened by more than a factor 2.4, theoretically supporting sub 60 fs pulses (cf. Fig. 8(a)&(b)). This shows that in addition to XPW, other nonlinear effects like self phase modulation are present. There is no compression stage installed after the XPW filter, so the compressed pulse duration of the XPW filtered pulse could not be measured. There are no modulations left in the spectral intensity, indicating that the post-pulses are sufficiently filtered by the XPW-stage. After the XPW, the divergence and size of the beam is adapted to match the cavity mode of the second Polaris amplifier A2 .
3. Temporal characteristics
3.1. Pulse shortening
After stretching and further amplification the pulses are compressed after the amplifier A2 or after A4 by a tiled grating compressor [2, 21] which completes the second CPA-stage. Without the spectral broadening of the XPW filter the pulses can be compressed down to 130 fs (A2, 15 mJ) or ≥160 fs (A4, 4 J). As the spectrum is broadened by the XPW filter, the pulses can be now compressed below 120 fs (A2) and 140 fs for the final amplifier (see Fig. 9(a)), resulting in a 15 % increased intensity in the focal spot. Note that the reduction of the pulse duration due to the broadened XPW spectrum is limited by spectral losses in the subsequent amplification chain.
Figure 9(b) shows a histogram of the duration of 30 consecutive shots of the final Polaris amplifier A4 at full energy operation measured during an experimental campaign, with 73 % of the pulses being shorter than 152 fs.
3.2. Temporal intensity contrast
Figure 10(a) shows the third order autocorrelation traces (Sequoia, Amplitude Technologies) of the filtered and unfiltered signals, measured with the pulses being amplified to the mJ level with the second amplifier A2. The black curve shows the temporal intensity contrast without the DCPA stage, having an ASE level of 10−9. In comparison, the green line gives the trace of the frontend with the DCPA but without the XPW filter. One can see two main differences here. First the ASE pedestal is about one order of magnitude higher, which is due to the smaller seed energy for the amplifier A1, since the pulse passes the ps-stretcher of the DCPA stage before it is being amplified. Second, as there is a waveplate in the amplifier, several pre- and post-pulses appear next to the main pulse (see inset Fig. 10(a)). These pulses can be avoided by replacing the waveplate by an all-reflective polarization rotator . However, these side pulses can effectively demonstrate the potential of the XPW filter.
The red line in Fig. 10 is the measurement performed with the XPW filtered pulse. The aforementioned side pulses are suppressed below the coherent pedestal of the main pulse, the remaining postpulses are generated by transmittive optics after the XPW stage. The relative temporal contrast of the filtered pulse approaches 10−10 at around −50 ps before the main pulse, reaching the detection limit of the correlator, resulting in noise (hatched area 10). Assuming this level as the ASE pedestal, an improvement of only 2 orders of magnitude would have been achieved. As the polarization contrast and efficiency of the XPW would indicate at least four orders of magnitude, we expected the ASE pedestal to be in the range 10−12 or less.
To verify this hypothesis, we blocked the seed of the amplifier to measure the ASE level of the complete amplifier chain with a photodiode (for a full ASE characterization of the laser before the update, see  for details). This measurement yields an ASE pedestal with a temporal intensity contrast of 2×10−13, indicated by the blue line. This is the limit set by the ASE generated by the amplifier A2 . To ultimately verify this result, we decreased the seed of the amplifier by a factor of 770 with an ND filter and increased the number of roundtrips to achieve the same output energy. As the amplifiers of Polaris are not working in saturation, the amplified spontaneous emission is increased by the same factor by which the seed is reduced . Thus the ASE pedestal of the pulses is expected to be around 10−10, what is just within the detection limit of the third order correlator.
In Figure 10, the green circles show an ASE level at the noise limit of the detector, which is the expected range. This proves that the XPW filter indeed reduces the ASE pedestal to a relative intensity level of 2×10−13. As the pulse’s leading edge can not be temporally resolved below 10−10 it is not verifiable, at which time before the main pulse the intensity reaches this level. Thus, the improvement of the temporal contrast in comparison to the frontend without the DCPA is four orders of magnitude, where the height of the final ASE pedestal is determined by the second amplifier of Polaris that follows the XPW-stage. For the fully amplified pulse the total energy of the amplified spontaneous emission is as small as 38 nJ. Further amplification of the pulse by the multipass amplifiers does not increase the ASE level, as the millisecond ASE by these amplifiers is at a relative intensity of 1.5×10−17 .
In this paper we have characterized a new frontend for the high-power, near-infrared, diode-pumped solid state laser system Polaris. To improve the intensity contrast of the laser system a second, picosecond CPA-stage was set up around the frontend amplifier. The pulses centered at a wavelength of 1030 nm are stretched, amplified and recompressed to 2 mJ energy and 112 fs duration.
A cross-polarized wave generation stage was used to temporally filter the pulses before they were further amplified, delivering more than 150 μJ pulse energy. The smooth Gaussian beam profile of the amplified pulses allowed for an internal conversion efficiency of the XPW process of more than 13 % in a single crystal setup without the need of spatial filtering. The challenge of achieving a high polarization contrast at the XPW stage including the compressor was overcome by using zero-phase-shift mirrors and precise adjustments of the used polarizers.
The reduced relative intensity level of the amplified spontaneous emission reached a level of 2×10−13 after the amplification of the pulses in the full amplifier chain of Polaris. This ASE level, to the best of our knowledge, is the lowest achieved so far with an all-diode pumped solid state laser system delivering high-energy femtosecond pulses. The spectral broadening by the XPW process lead to a reduced pulse duration of the fully amplified pulses from 164 fs to 144 fs, resulting in an 15 % increased intensity in the focal spot.
The excellent performance of the presented frontend shows its capability delivering high-quality seed pulses for high-intensity laser systems, such as Polaris.
The research leading to these results has received funding from the European commission’s (EC) 7th Framework Programme (LASERLAB-EUROPE, grant no. 228334) and from the Bundesministerium für Bildung und Forschung (BMBF) ( 03ZIK445 and 03Z1H531).
References and links
1. D. Strickland and G. Mourou, “Compression of amplified chirped optical pulses,” Opt. Commun. 55, 447–449 (1985). [CrossRef]
2. M. Hornung, H. Liebetrau, A. Seidel, S. Keppler, A. Kessler, J. Körner, M. Hellwing, F. Schorcht, D. Klöpfel, A. K. Arunachalam, G. Becker, A. Sävert, J. Polz, J. Hein, and M. C. Kaluza, “The all-diode pumped laser system Polaris - an experimentalist’s tool generating ultra-high contrast pulses with high energy,” High Power Laser Sci. Eng. 2, e20 (2014).
3. N. Hopps, C. Danson, S. Duffield, D. Egan, S. Elsmere, M. Girling, E. Harvey, D. Hillier, M. Norman, S. Parker, P. Treadwell, D. Winter, and T. Bett, “Overview of laser systems for the Orion facility at the AWE,” Appl. Opt. 52, 3597–3607 (2013). [CrossRef] [PubMed]
4. S. Fourmaux, S. Payeur, S. Buffechoux, P. Lassonde, C. St-Pierre, F. Martin, and J. C. Kieffer, “Pedestal cleaning for high laser pulse contrast ratio with a 100 TW class laser system,” Opt. Express 19, 8486–8497 (2011). [CrossRef] [PubMed]
5. J. Liu, K. Okamura, Y. Kida, and T. Kobayashi, “Temporal contrast enhancement of femtosecond pulses by a self-diffraction process in a bulk Kerr medium,” Opt. Express 18, 22245–22254 (2010). [CrossRef] [PubMed]
6. C. Dorrer, I. A. Begishev, A. V. Okishev, and J. D. Zuegel, “High-contrast optical-parametric amplifier as a front end of high-power laser systems,” Opt. Lett. 32, 2143–2145 (2007). [CrossRef] [PubMed]
7. F. Wagner, C. P. Joao, J. Fils, T. Gottschall, J. Hein, J. Körner, J. Limpert, M. Roth, T. Stöhlker, and V. Bagnoud, “Temporal contrast control at the PHELIX petawatt laser facility by means of tunable sub-picosecond optical parametric amplification,” Appl. Phys. B-Lasers Opt. 116, 429–435, (2014) [CrossRef]
8. D. Hillier, C. Danson, S. Duffield, D. Egan, S. Elsmere, M. Girling, E. Harvey, N. Hopps, M. Norman, S. Parker, P. Treadwell, D. Winter, and T. Bett, “Ultrahigh contrast from a frequency-doubled chirped-pulse-amplification beamline,” Appl. Opt. 52, 4258–4263 (2013). [CrossRef] [PubMed]
9. G. I. Petrov, O. Albert, J. Etchepare, and S. M. Saltiel, “Cross-polarized wave generation by effective cubic nonlinear optical interaction,” Opt. Lett. 26, 355–357 (2001). [CrossRef]
11. A. Jullien, O. Albert, F. Burgy, G. Hamoniaux, L. P. Rousseau, J. P. Chambaret, F. Auge-Rochereau, G. Cheriaux, J. Etchepare, N. Minkovski, and S. M. Saltiel, “10(−10) temporal contrast for femtosecond ultraintense lasers by cross-polarized wave generation,” Opt. Lett. 30, 920–922 (2005). [CrossRef] [PubMed]
12. A. Cotel, A. Jullien, N. Forget, O. Albert, C. Le Blanc, J. Etchepare, G. Cheriaux, N. Minkovski, and S. M. Saltiel, “Nonlinear temporal pulse cleaning for high-energy petawatt-class lasers,” J. Phys. IV 133, 713–715 (2006).
13. A. Jullien, L. Canova, O. Albert, D. Boschetto, L. Antonucci, Y. H. Cha, J. P. Rousseau, P. Chaudet, G. Cheriaux, J. Etchepare, S. Kourtev, N. Minkovski, and S. M. Saltiel, “Spectral broadening and pulse duration reduction during cross-polarized wave generation: influence of the quadratic spectral phase,” Appl. Phys. B-Lasers Opt. 87, 595–601 (2007). [CrossRef]
14. M. P. Kalashnikov, E. Risse, H. Schonnagel, A. Husakou, J. Herrmann, and W. Sandner, “Characterization of a nonlinear filter for the front-end of a high contrast double-cpa Ti : sapphire laser,” Opt. Express 12, 5088–5097 (2004). [CrossRef] [PubMed]
15. Y. Zaouter, L. P. Ramirez, D. N. Papadopoulos, C. Honninger, M. Hanna, F. Druon, E. Mottay, and P. Georges, “Temporal cleaning of a high-energy fiber-based ultrafast laser using cross-polarized wave generation,” Opt. Lett. 36, 1830–1832 (2011). [CrossRef] [PubMed]
16. A. Cotel, A. Jullien, N. Forget, O. Albert, G. Cheriaux, and C. Le Blanc, “Nonlinear temporal pulse cleaning of a 1 μm optical parametric chirped-pulse amplification system,” Appl. Phys. B-Lasers Opt. 83, 7–10 (2006). [CrossRef]
17. M. Hornung, S. Keppler, R. Bödefeld, A. Kessler, H. Liebetrau, J. Körner, M. Hellwing, F. Schorcht, O. Jäckel, A. Sävert, J. Polz, A. K. Arunachalam, J. Hein, and M. C. Kaluza, “High-intensity, high-contrast laser pulses generated from the fully diode-pumped Yb:glass laser system polaris,” Opt. Lett. 38, 718–720 (2013). [CrossRef] [PubMed]
18. S. Keppler, R. Bödefeld, M. Hornung, A. Sävert, J. Hein, and M. C. Kaluza, “Prepulse suppression in a multi-10-TW diode-pumped Yb:glass laser,” Appl. Phys. B-Lasers Opt. 104, 11–16 (2011). [CrossRef]
19. L. P. Ramirez, D. Papadopoulos, M. Hanna, A. Pellegrina, F. Friebel, P. Georges, and F. Druon, “Compact, simple, and robust cross polarized wave generation source of few-cycle, high-contrast pulses for seeding petawatt-class laser systems,” J. Opt. Soc. Am. B 30, 2607–2614 (2013). [CrossRef]
20. A. Jullien, J. P. Rousseau, B. Mercier, L. Antonucci, O. Albert, G. Cheriaux, S. Kourtev, N. Minkovski, and S. M. Saltiel, “Highly efficient nonlinear filter for femtosecond pulse contrast enhancement and pulse shortening,” Opt. Lett. 33, 2353–2355 (2008) [CrossRef] [PubMed]
21. M. Hornung, R. Bödefeld, M. Siebold, A. Kessler, M. Schnepp, R. Wachs, A. Sävert, S. Podleska, S. Keppler, J. Hein, and M. C. Kaluza, “Temporal pulse control of a multi-10 TW diode-pumped Yb:glass laser,” Appl. Phys. B-Lasers Opt. 101, 93–102 (2010). [CrossRef]
22. S. Keppler, M. Hornung, R. Bödefeld, M. Kahle, J. Hein, and M. C. Kaluza, “All-reflective, highly accurate polarization rotator for high-power short-pulse laser systems,” Opt. Express 20, 20742–20747 (2012). [CrossRef] [PubMed]
23. S. Keppler, M. Hornung, R. Bödefeld, A. Sävert, H. Liebetrau, J. Hein, and M. C. Kaluza, “Full characterization of the amplified spontaneous emission from a diode-pumped high power laser system,” Opt. Express 22, 11228–11235 (2014). [CrossRef] [PubMed]