We report on a broadband OPCPA system, pumped at 515 nm by frequency doubled Yb:YAG thin disk lasers. The system delivers 11.3 mJ pulses at a central wavelength of 800 nm with a spatial beam quality of M2 = 1.25 and > 25% pump-to-signal conversion efficiency. The broadband pulses were demonstrated to be compressible to 12 fs using a chirped mirror compressor.
© 2016 Optical Society of America
Ultra-short pulse trains with both a high repetition rate (1 kHz and higher) and multi-mJ pulse energy are desirable in the fields of material science, soft X-ray spectroscopy, and molecular science [1–3]. Currently, the primary methods of producing such pulse trains are direct amplification in Ti:Sapphire laser systems  and optical parametric chirped pulse amplification (OPCPA) [5,6]. OPCPA offers significant potential advantages for producing high energy, high average power pulse trains. Because the amplification is a parametric process, there is no energy stored in the amplifying medium, thus reducing heating effects and allowing for amplification to high average powers. OPCPA, especially in the non-collinear configuration, offers an exceptionally large gain bandwidth which can even be slightly tuned with the angle of the nonlinear crystal . Hence, amplification to very high energies is achievable without significant bandwidth limiting due to gain narrowing. Additionally, there is no amplified spontaneous emission in OPCPA and, outside the temporal window of the pump pulse, there is no mechanism for light emission. This in turn improves pulse contrast at the target, especially in the case of OPCPA pumped by picosecond pulses.
The Extreme Light Infrastructure (ELI) – Beamlines facility is constructing a high repetition rate ultra-short pulse laser system for user-based research with the following target parameters: 100 mJ pulse energy, 1 kHz repetition rate, and 20 fs pulse duration centered at 820 nm. The primary technology used in constructing this beamline is Yb:YAG thin disk-pumped picosecond OPCPA. Thin disk lasers have proved to be a very reliable method of amplifying pulses to high energies at high repetition rates with exceptional beam quality [8,9] and thin disk pumped OPCPA has been successfully demonstrated at a variety of repetition rates [10,11].
In the design of this high repetition rate beamline, compression of the high power pulses is intended to be done with chirped mirrors, while fine dispersion tuning is performed before amplification using an acousto-optical programmable dispersive filter (AOPDF). This allows for good scalability to higher energies and also provides flexibility in the dispersion management. The group delay dispersion (GDD) per bounce from a chirped mirror supporting the pulse bandwidth is typically on the order of −120 fs2. To compress the positively chirped broadband pulses, which are stretched to 1.5 ps, roughly 40 bounces on such mirrors is required. Producing large aperture, GDD ripple-free chirped mirror pairs with high GDD per bounce for the 720–950 nm broadband spectrum is very challenging. For this reason, pulse compressibility tests were performed to evaluate the effect of seed pulses with pre-formed phase on the amplification in the OPCPA chain in terms of efficiency and final output spectrum.
In this work we report on the performance of the laser beamline front end which produces > 11 mJ pulses at 1 kHz via picosecond OPCPA and demonstrate high quality compression using an AOPDF in combination with chirped mirrors. It is worth emphasizing that this laser is a front end for a larger beamline. Hence, the pulses at this stage are intended to remain chirped as they pass to the main beamline. For this reason, full power compression is not a priority for this study. The focus lies on chirped pulse energy, beam quality, and pulse compressibility.
2. Experimental setup
A schematic setup of the experiment is shown in Fig. 1. The system consists of a dual output broadband Ti:sapphire oscillator (Femtolasers, Rainbow), two Yb:YAG thin disk regenerative amplifiers, and a four-stage OPCPA amplification chain. The Ti:sapphire oscillator, operating at 80 MHz, produces a seed for both pump and signal chains, assuring good initial synchronization, which is necessary for picosecond OPCPA. The broadband output of the oscillator gives 2 nJ pulses at 800 nm with a duration of 6 fs, whereas the second output is centered at a 1030 nm and is used as a seed for the pump lasers.
The 1030 nm seed is coupled into an all fiber based seed conditioning setup for the Yb:YAG pump lasers. There, the 1030 nm output is stretched in chirped fiber Bragg gratings, pulse-picked down to 1 kHz repetition rate, amplified using Yb:doped fiber amplifiers, and finally coupled into two Yb:YAG thin disk-based regenerative amplifiers (laser heads from TRUMPF Scientific Lasers), producing 26 mJ and 69 mJ of compressed 1030 nm pulses at 1 kHz. The pump beams are compressed in multi-layer dielectric (MLD) grating compressors (gratings by Plymouth Grating Laboratory) to 3.0 ps and 2.2 ps, respectively, frequency doubled in LBO crystals to the 515 nm OPCPA pump wavelength with a yield of 12.5 mJ and 32 mJ respectively, and sent to the OPCPA amplifier. More details on pump laser systems can be found in [9,12].
The broadband OPCPA chain consists of an AOPDF (Dazzler HR45-650-1100, Fastlite), a three-stage OPCPA preamplifier using BBO crystals, followed by an OPCPA stage using LBO crystal, and a downscaled chirped mirror compressor for demonstration of compressibility of the amplified signal. The AOPDF positively stretches the broadband seed with an approximate average GDD of +5000 fs2 and compensates for higher order dispersion of the system. The pump energy is distributed to the OPCPA stages in such a way as to keep the peak intensity below 100 GW/cm2 in order to prevent optical damage of the crystals, while also maximizing the efficiency of the main amplifier. The detailed parameters of the OPCPA system are presented in Table 1. To reduce the effects of self-focusing on the pump beam as it propagates though air, the beam path is minimized. Additionally, the pump beam is expanded to 5 mm (FWHM) between the SHG and the third stage of OPCPA, and then matched to a desired size on the OPA crystal according to Table 1.
The thin disk regenerative amplifiers have exceptionally good beam quality , which we exploit in the OPCPA. In order to maintain the Gaussian shape of the signal beam, we chose to have it slightly larger than the pump, which acts now as a soft aperture for the signal at each stage. The OPCPA stages use non-collinear type I phase matching, with the first two stages phase matched in walk-off compensating configuration, providing better OPCPA efficiency for narrow beams; the other stages are in walk-off non-compensating configuration, which suffers less from a parasitic second harmonic generation of the signal and idler. The orientation of crystals is described in detail in Table 1.
A small portion of about 100 µJ of the amplified beam after the last stage with a diameter of 5 mm is compressed in a chirped mirror compressor (CMC). The compressor consists of two 3 inch-diameter chirped mirrors and 8 half-inch steering silver mirrors, allowing for 42 bounces on the chirped mirrors. The 3 inch diameter is used here because it is the aperture required for compression of the future 100 mJ laser system. Therefore, the CMC gives us a direct test of the performance of these large aperture mirrors. The chirped mirrors have an average GDD per bounce of −120 fs2 in the wavelength range of 710–930 nm and a reflectivity of > 99.8%. Thus, the expected throughput of the final compressor is roughly 90%. The chirped mirror pair was selected from a number of chirped mirrors from different suppliers based on a dispersion measurement and a numerical evaluation of the possible pulse shape after compression. Additionally, we have performed a uniformity test of the large aperture mirrors in terms of their GDD. The results show there is a small unidirectional shift in the GDD curves across the mirror. They preserve their shape, however they are shifted by approximately ±10 nm compared to the curve measured in the center. This can be compensated to some extent by careful orientation of the chirped mirrors inside the compressor. The collaboration with chirped mirror manufacturers is still in progress to achieve better chirped mirror characteristics.
To assure high stability and long term reliability of the high average power OPCPA system pumped by picosecond pulses, the following aspects have to be addressed. First of all, the temporal synchronization of the pump and signal is assured by jitter stabilization modules, consisting of an optical balanced cross-correlator . The stabilization modules provide feedback for the delay control of the pump lasers and synchronize the pump pulse with the signal to a level of 15 fs RMS. Additionally, the regenerative amplifier 1 is equipped with a beam pointing stabilization system based on piezo actuated mirrors, reducing the pointing errors to below 1.2µrad RMS and centering error to < 6.5µm RMS. The actively controlled spatial overlap is critical for stable amplification in the OPCPA stages pumped with narrow beams and also improves the beam pointing stability of the whole amplification chain.
3. Performance of the laser system
The output of the whole 4-stage amplification chain is 11.3 mJ, which corresponds to a total pump-to-signal efficiency of 25% with > 28% efficiency in the last OPCPA stage, which is similar to what was reported in . The relative energy stability was measured to be 3.1% RMS as shown in Fig. 2.
In Fig. 3(a) we present the spectrum of the amplified signal, which spans the wavelength range of 730–940 nm. Although the BBO-based OPA in general supports a broader amplification bandwidth, the spectrum of the signal is intentionally limited to the 730 nm on the blue edge since LBO crystals are used in further OPA stages. The last three amplification stages are intended to be under vacuum, and the idler absorption would lead to non-negligible heating of the LBO crystals. Therefore, the optics in the system were chosen for that specific bandwidth and corresponding settings were applied to the AOPDF.
The shape of the amplified spectrum depends on the phase profile applied to the seed by the Dazzler. In case of a pure polynomial phase with nonzero terms up to TOD the achieved signal energy is 12.4 mJ and the amplified spectrum is nicely shaped, as shown in Fig. 3(a). However, a more complex phase, matching the higher order dispersion of the CMC, results in approximately 10% lower signal energy and noticeable spectral distortions as seen by the green line in Fig. 3(a). This is a result of not equally stretched wavelengths in OPCPA. The higher order phase is shown as red line in Fig. 3(b). Despite the modulated spectral intensity, the measured temporal profile of the compressed pulse was measured to be close to the Fourier limited pulse for the given spectrum, as shown in Fig. 3(c).
The spatial quality of the amplified broadband beam is presented in Fig. 4. The M2 was measured to be 1.22 and 1.25 along the x and y axes. For the derivation of the M2 we use a wavelength of 797 nm, which is a central wavelength weighted by spectral intensity shown in Fig 3(a). The spatial profile of the amplified beam is close to Gaussian, with a narrow plateau around its peak, which is caused by saturation of parametric amplification in the last stage of OPCPA. The spatial profile of amplified beam is shown in Fig. 4(b).
We have reported on the generation of 11.3 mJ broadband pulses delivered by the OPCPA system pumped by frequency doubled Yb:YAG thin disk lasers. The output has good spatial quality and is compressible to near FT-limited 12.1 fs. The system presented here is a completed front end, which paves the way for the development of the main beamline. Most importantly, the presented work provides a proof of principle of the beamline concept, namely all-picosecond OPCPA broadband amplification and all-chirped mirror compression. While chirped mirror compression is highly efficient and scalable to high energies, production of GDD ripple-free chirped mirrors with both large chirp and broad spectral bandwidth remains challenging. Demonstrating that an off-the-shelf AOPDF is able to compensate the complicated residual GDD ripple from the chirped mirrors to produce high quality femtosecond pulses is an important validation of this type of laser design. In the next step, another three OPCPA stages and a fully scaled CMC will be implemented in vacuum to produce multi-TW peak-power from 100 mJ, <20 fs pulses, providing users with a powerful tool in various fields of science.
Extreme Light Infrastructure (CZ.02.1.01/0.0/0.0/15 008/0000162), CITT (CZ.1.05/3.1.00/10.0210), Czech Ministry of Education, Youth and Sports (RVO 68407700).
The authors are grateful to Fastlite, Inc. for their assistance in this work.
References and links
1. K. Sugioka and Y. Cheng, “Ultrafast lasers – reliable tools for advanced materials processing,” Light Sci. Appl. 3(4), e149 (2014). [CrossRef]
2. E. Gagnon, P. Ranitovic, X. M. Tong, C. L. Cocke, M. M. Murnane, H. C. Kapteyn, and A. S. Sandhu, “Soft X-ray-driven femtosecond molecular dynamics,” Science 317(5843), 1374–1378 (2007). [CrossRef] [PubMed]
3. H. J. Wörner, J. B. Bertrand, D. V. Kartashov, P. B. Corkum, and D. M. Villeneuve, “Following a chemical reaction using high-harmonic interferometry,” Nature 466(7306), 604–607 (2010). [CrossRef] [PubMed]
4. A. Jullien, A. Ricci, F. Böhle, J. P. Rousseau, S. Grabielle, N. Forget, H. Jacqmin, B. Mercier, and R. Lopez-Martens, “Carrier envelope-phase stable, high-contrast, double chirped-pulse-amplification laser system,” Opt. Lett. 39(13), 3774–3777 (2014). [CrossRef] [PubMed]
5. S. Adachi, N. Ishii, T. Kanai, A. Kosuge, J. Itatani, Y. Kobayashi, D. Yoshitomi, K. Torizuka, and S. Watanabe, “5-fs, multi-mJ, CEP-locked parametric chirped-pulse amplifier pumped by a 450-nm source at 1 kHz,” Opt. Express 16(19), 14341–14352 (2008). [CrossRef] [PubMed]
6. K. H. Hong, C. J. Lai, J. P. Siqueira, P. Krogen, J. Moses, C. L. Chang, G. J. Stein, L. E. Zapata, and F. X. Kärtner, “Multi-mJ, kHz, 2.1µm optical parametric chirped-pulse amplifier and high-flux soft X-ray high-harmonic generation,” Opt. Lett. 39(11), 3145–3148 (2014). [CrossRef] [PubMed]
7. G. Cerullo and S. De Silvestri, “Ultrafast optical parametric amplifiers,” Rev. Sci. Instrum. 74(1), 1–18 (2003). [CrossRef]
8. T. Metzger, A. Schwarz, C. Y. Teisset, D. Sutter, A. Killi, R. Kienberger, and F. Krausz, “High-repetition-rate picosecond pump laser based on a Yb:YAG disk amplifier for optical parametric amplification,” Opt. Lett. 34(14), 2123–2125 (2009). [CrossRef] [PubMed]
9. J. Novák, J. T. Green, T. Metzger, T. Mazanec, B. Himmel, M. Horáček, R. Boge, R. Antipenkov, F. Batysta, J. A. Naylon, P. Bakule, and B. Rus, “Thin disk amplifier-based 40 mJ, 1 kHz, picosecond laser at 515 nm,” Opt. Express 24, 5728–5733 (2016). [CrossRef] [PubMed]
10. S. Prinz, M. Haefner, C. Y. Teisset, R. Bessing, K. Michel, Y. Lee, X. T. Geng, S. Kim, D. E. Kim, T. Metzger, and M. Schultze, “CEP-stable, sub-6 fs, 300-kHz OPCPA system with more than 15 W of average power,” Opt. Express 23(2), 1388–1394 (2015). [CrossRef] [PubMed]
11. H. Fattahi, H. G. Barros, M. Gorjan, T. Nubbemeyer, B. Alsaif, C. Y. Teisset, M. Schultze, S. Prinz, M. Haefner, M. Ueffing, A. Alismail, L. Vámos, A. Schwarz, O. Pronin, J. Brons, X. T. Geng, G. Arisholm, M. Ciappina, V. S. Yakovlev, D. E. Kim, A. M. Azzeer, N. Karpowicz, D. Sutter, Zs. Major, T. Metzger, and F. Krausz, “Third-generation femtosecond technology,” Optica 1(1), 45–63 (2014). [CrossRef]
12. J. Novák, P. Bakule, J. T. Green, F. Batysta, T. Metzger, J. Hřebíček, J. A. Naylon, T. Mazanec, M. Vítek, and B. Rus, “Thin disk picosecond pump laser for jitter stabilized kHz OPCPA,” Proc. SPIE 8780, 878020 (2013). [CrossRef]
13. A. V. Smith, “How to select nonlinear crystals and model their performance using SNLO software,” Proc. SPIE 3928, 62–69 (2000). [CrossRef]
14. F. Batysta, R. Antipenkov, J. T. Green, J. A. Naylon, J. Novák, T. Mazanec, P. Hříbek, Ch. Zervos, P. Bakule, and B. Rus, “Pulse synchronization system for picosecond pulse-pumped OPCPA with femtosecond-level relative timing jitter,” Opt. Express 22, 30281–30287 (2014). [CrossRef]
15. S. Prinz, M. Haefner, C. Y. Teisset, R. Bessing, K. Michel, Y. Lee, X. T. Geng, S. Kim, D. E. Kim, T. Metzger, and M. Schultze, “CEP-stable, sub-6 fs, 300-kHz OPCPA system with more than 15 W of average power,” Opt. Express 23, 1388–1394 (2014). [CrossRef]