We present a two-color pumped OPCPA system which delivers an ultra-broadband spectrum spanning from 430 nm to 1.3 µm with a Fourier limited pulse duration of sub-3 fs and 1 µJ of pulse energy at a repetition rate of 200 kHz. All frequency components propagate on a common path, thus the spectral phase along the whole spectrum is well-defined. The inner part of the spectrum has been compressed to sub-5 fs pulses.
© 2012 OSA
Currently, optical parametric chirped pulse amplifiers (OPCPA) are the most promising ultra-broadband amplification concepts with respect to energy scaling. In non-collinear geometry the amplification bandwidth can be enhanced up to the few-cycle regime, and pulse durations of 5.5 fs around 800 nm from a 1 kHz system (2.7 mJ) have been demonstrated  as well as sub-5 fs around 650 nm . Both groups could further reduce the pulse durations by spectral broadening in gaseous media and subsequently pulse compression (sub-4 fs, 1.2 mJ)  and/or using adaptive optics for pulse compression (4 fs, 0.5 µJ) . Also with higher repetition rates (35 kHz) sub-5 fs pulses from a fiber-amplifer-pumped OPCPA system have been realized by . Schultze et al. demonstrated a compact OPCPA system working with repetition rates up to 500 kHz delivering carrier-envelope phase stabilized two-cycle pulses around 800 nm with multi-µJ of pulse energy .
Beyond that, research is conducted to reach even shorter pulse durations by combining complemental spectra into one single-cycle pulse [7–10] or to broaden the amplification bandwidth using two subsequent optical parametric amplification (OPA) stages with different pump wavelengths. In  a green pumped OPA stage for the NIR part is followed by a blue pumped stage amplifying the VIS part of an ultra-broadband seed spectrum. The output spectrum supports a Fourier limited pulse duration of 4.5 fs with a repetition rate of 10 Hz; compression was not demonstrated. In a second experiment, Hermann et al. proved the compressibility of a much narrower spectrum (FL: 11 fs) generated by the same two-color amplification scheme with a 100 kHz system down to 13 fs .
Here, we report on our two-color pumped double-stage OPCPA system operating at a repetition rate of 200 kHz. The seed is generated by an octave-spanning Ti:sapphire oscillator (optionally CEP stabilized). After a first amplification stage (Fig. 1 : SH-NOPA) the spectral bandwidth is extended to the blue wavelength region by white-light generation in a bulk material (Fig. 1: WLG). A second OPA stage amplifies the visible part while the infrared is transmitted nearly unchanged in the same beam (Fig. 1: TH-NOPA). The pump pulses for the two NOPA stages are the second and third harmonic from a regenerative thin-disk amplifier (tuneable repetition rate from 100 kHz to 500 kHz), seeded by the infrared part of the Ti:sapphier oscillator. This compact setup delivers ultra-broadband spectra supporting pulses in the single-cycle regime with 1 µJ of pulse energy at high repetition rates.
2. Experimental setup and results
The front-end of the experimental setup (see Fig. 2 ) consists of a commercial Ti:sapphire oscillator (VENTEON | PULSE ONE) which delivers both, the broadband seed for the OPA stages, and the narrowband seed around 1030 nm for the Yb:YAG pump amplifier.
Second and third harmonic generation
The infrared seed from the Ti:sapphire oscillator is amplified in the home-built Yb:YAG thin-disk regenerative amplifier to 35 µJ of pulse energy with a repetition rate of 200 kHz and pulse durations of 1.5 ps . A collinear geometry is used to convert the fundamental 1030 nm radiation to its second (SH; 515 nm) and subsequently to its third harmonic (TH; 343 nm) by sum frequency generation. In this respect, the output of the regenerative amplifier is weakly focused (f = 500 mm) down into an ~800 µm long type-I BBO (θ = 23.4°) crystal followed directly by a 2 mm long type-II LBO crystal (θ = 50.1°, φ = 90°). Details about this simple and efficient set-up can be found in . The distance between the two crystals is approximately 135 mm, where the BBO is placed 35 mm in front of the focus (focus diameter: 130 µm) and the LBO 100 mm behind the focus to avoid damages and to ensure a stable operation. The ratio between the SH and TH output power can be easily adjusted by tuning the phase matching angle of the LBO crystal. In operation we use 10 µJ of SH as pump for the first NOPA stage and 7 µJ of TH for the second stage. Two dichroic beam splitters separate the three beams for independent collimation.
Second harmonic pumped NOPA
The main part from the Ti:sapphire oscillator is stretched by two BK7 wedges to a pulse duration of almost 150 fs and subsequently amplified in the first NOPA stage, where it is non-collinearly superimposed with the SH pump in a 5 mm thick type-I BBO crystal (θp ≈24°). The BBO dispersion further stretches the seed pulse to nearly 400 fs. The focal diameters of the seed and pump beams are about 100 µm. The internal non-collinear angle is close to 2.4°, which enables a broadband amplification of the seed spectrum. Figure 3(a) shows the seed spectrum (black dashed curve) as well as the amplified signal spectrum (red solid curve), the difference between the two spectra reveals the broadband amplification. The main structures of the seed are conserved during the amplification. As a result of the non-collinear walk-off compensation geometry, where the pump beam is positioned between the optical axis of the uniaxial BBO crystal and the seed beam, a parasitic SH effect is phase matched in a spectral region of the amplified signal . In Fig. 3(a) the dip around 340 THz can be attributed to this SHG. Please note that the pump pulses come with a repetition rate of 200 kHz while the seed is not picked; so, every 381th seed pulse is amplified. As a consequence, the averaged measurements, e.g. of power or power spectra, have to be corrected with respect to the non-amplified background. With this correction the first NOPA stage delivers pulses with 1 µJ of pulse energy in broadband spectra supporting pulse durations below 7 fs.
For the supercontinuum generation the input beam profile is of major importance and an aperture can be placed in front of the amplification stage in order to clean the beam. This of course reduces the pulse energy, but leads to a more stable white-light generation. The signal behind the first NOPA stage is compressed using 16 bounces on dispersive mirrors and focused down by a concave mirror (ROC = −200 mm) into a 3 mm long BaF2 plate. The compressed pulse duration inside the plate is below 20 fs. BaF2 is located between sapphire and CaF2 with respect to the characteristic power for self-focusing . Concerning the stability and seed spectrum inside the spectral amplification window of the second NOPA, we choose BaF2 for white-light generation. During operation, the plate has to be slowly moved back and forth perpendicular to the beam to avoid irreversible defects in the crystal. The output spectrum spans from 430 nm to 1300 nm at – 40 dB. In the crystal, only 30 to 40 percent of the pulse energy propagates in the filament core, the rest is allocated in the reservoir of the filament and lost for further experiments; less than 0.2 percent from the incident power is converted to the visible spectral part. A photograph of the white-light is shown in Fig. 3(b). Discrimination of the reservoir from the filament core is achieved using apertures at different locations behind the BaF2 crystal.
Third harmonic pumped NOPA
The second NOPA stage amplifies the visible components of the supercontinuum seed. The TH (7 µJ) is focused down to a diameter of approximately 180 µm into a 5 mm thick type-I BBO crystal (θp ≈37.2 °). The white-light seed is focused down by a concave mirror (ROC=−800 mm) into the BBO, where the internal non-collinear angle between pump and seed is chosen to be α ≈4.5 °. Temporal stretching of the seed pulse by adding further dispersion in front of the BBO was not beneficial. The phase matching condition determines the amplification of the visible part, whereas the remaining spectrum passes nearly unchanged. Careful alignment of pump and seed beam with respect to θp, α, and the beam radius is necessary to keep the output in a single beam in the near- and in the far field.
Figure 4(a) shows the white-light seed spectrum (black dashed curve) and the resulting signal output spectrum (red solid curve). The difference between the two spectra emphasizes the broadband amplification bandwidth from 430 nm to 700 nm of the second NOPA stage. The whole spectral bandwidth spans from 230 THz to 690 THz (at −40 dB), which is equivalent to 1.5 optical octaves. The final spectrum shows a prominent peak around the central frequency at 450 THz which is due to phase matching in the second OPA; nevertheless, the Fourier limited pulse duration is 2.5 fs which corresponds to 1.2 optical cycles (see Fig. 4(b)) with 83% of energy in the main peak. The pulses contain 1 µJ of pulse energy at a repetition rate of 200 kHz. Since all spectral components are generated and propagate in one single beam, the spectral phase along the whole ultra broadband spectra is well defined, allowing in principle for compression with a static dispersion compensation scheme.
We employed 23 bounces on octave spanning double chirped mirrors (500 to 1000 nm, GDD of −100 fs2 for the mirror pair @ 700 nm) and some wedges for dispersion control and performed a SPIDER measurement of this inner part of the spectrum. With a spectral filter (OG 530) in front of the SPIDER setup, we cut all fundamental wavelengths below 530 nm.
The resulting spectrum (solid red curve in Fig. 5(a) ) still keeps a Fourier limited pulse duration of 4 fs. Figure 5 shows the measured spectral phase and the temporal pulse envelope with a FWHM of 4.6 fs.
The compression of the whole output spectrum needs a detailed knowledge of the spectral phase. Origin and generation as well as the amplification and propagation of the spectral components have to be taken into account. After the supercontinuum generation the full spectral bandwidth is present and all spectral components are included in the same beam. The main contributions to the final spectral phase curve are the seed and the crystal dispersion; both are well known. Whereas, the role of our NOPA phases and the phase from the white- light generation process are unexplored yet. Herrmann et al. published the calculated and measured phase contribution for a SH pumped NOPA stage followed by a TH pumped NOPA stage . The measured spectral phase – especially around the spectral overlap region of the two stages – showed a fairly flat behavior. Regarding the white-light generation in a bulk medium, Cerullo et al. mentioned a distortion of the spectral phase in the vicinity of the pump frequency [16, 17]. This is the reason why Antipenkov et al. and others commented that it is difficult to compress a full supercontinuum spectrum including the VIS and NIR part to its Fourier limit [18, 19]. Self-phase modulation strongly affects the spectral phase curve, leading to prominent phase jumps in the group delay of the driving spectrum of the WLG. Nevertheless, the magnitude of these jumps scales with the pulse duration. In  a 200 fs jump was measured with a WLG driving pulse duration of 150 fs. Here, with pulse durations below 20 fs this effect is much less critical and barely visible in the SPIDER measurement.
However, our SPIDER measurement of the reduced spectrum (Fig. 5(a), including the WLG pump wavelength) verifies that a simple DCM compression can flatten the phase. So, we are very confident that – especially with adaptive methods, e.g. a prism-based 4-f LCD pulse shaper setup  – it will be possible to control the spectral phase and compress the pulse to the single cycle. This, together with the pulse characterization, is subject of further investigations, as well as the energy scaling of the system, which is a question of the scalability of the white-light generation process and requires media with a higher critical power for self-focusing (e.g. CaF2, LiF2, gases).
In conclusion we reported on our compact high-repetition rate (200 kHz) two-color pumped OPCPA setup which allows for the generation of a coherent 450 THz broad spectrum centered at 650 nm with 1 µJ of pulse energy. The output spectrum supports a Fourier limited pulse duration of sub-3 fs corresponding to 1.2 optical cycles. The inner part of the spectrum has been compressed successfully to a pulse duration of 4.6 fs. This OPCPA system provides a unique source for spectroscopic applications, coherent control experiments, and fundamental studies in the single-cycle regime.
This work was partially supported by the Cluster of Excellence “Center of Quantum Engineering and Space-Time Research - QUEST”, founded by the German Research Foundation (DFG).
References and links
1. 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]
2. A. Shirakawa, I. Sakane, M. Takasaka, and T. Kobayashi, “Sub-5-fs visible pulse generation by pulse-front-matched noncollinear optical parametric amplification,” Appl. Phys. Lett. 74(16), 2268 (1999). [CrossRef]
4. A. Baltuška, T. Fuji, and T. Kobayashi, “Visible pulse compression to 4 fs by optical parametric amplification and programmable dispersion control,” Opt. Lett. 27(5), 306–308 (2002). [CrossRef] [PubMed]
5. S. Hädrich, S. Demmler, J. Rothhardt, C. Jocher, J. Limpert, and A. Tünnermann, “High-repetition-rate sub-5-fs pulses with 12 GW peak power from fiber-amplifier-pumped optical parametric chirped-pulse amplification,” Opt. Lett. 36(3), 313–315 (2011). [CrossRef] [PubMed]
6. M. Schultze, T. Binhammer, G. Palmer, M. Emons, T. Lang, and U. Morgner, “Multi-μJ, CEP-stabilized, two-cycle pulses from an OPCPA system with up to 500 kHz repetition rate,” Opt. Express 18(26), 27291–27297 (2010). [CrossRef] [PubMed]
7. R. Banici and D. Ursescu, “Spectral combination of ultrashort laser pulses,” EPL 94(4), 44002 (2011). [CrossRef]
8. G. Krauss, S. Lohss, T. Hanke, A. Sell, S. Eggert, R. Huber, and A. Leitenstorfer, “Synthesis of a single cycle of light with compact erbium-doped fibre technology,” Nat. Photonics 4(1), 33–36 (2010). [CrossRef]
9. S. Huang, G. Cirmi, J. Moses, K. Hong, S. Bhardwaj, J. R. Birge, L. Chen, E. Li, B. Eggleton, G. Cerullo, and F. X. Kärtner, “High-energy pulse synthesis with sub-cycle waveform control for strong-field physics,” Nat. Photonics 5(8), 475–479 (2011). [CrossRef]
10. S. Haessler, T. Balčiūnas, G. Andriukaitits, O. D. Mücke, A. Pugžlys, A. Baltuška, R. Squibb, L. Frasinski, J. Marangos, J. W. G. Tisch, L. Giniūnas, R. Danielius, and R. Holzwarth, “Multicolor optical parametric synthesizer for high-field science” in High Intensity Lasers and High Field Phenomena (HILAS), Technical Digest (CD) (Optical Society of America, 2011), paper HThD4.
11. D. Herrmann, C. Homann, R. Tautz, M. Scharrer, P. S. Russell, F. Krausz, L. Veisz, and E. Riedle, “Approaching the full octave: noncollinear optical parametric chirped pulse amplification with two-color pumping,” Opt. Express 18(18), 18752–18762 (2010). [CrossRef] [PubMed]
12. M. Schultze, T. Binhammer, A. Steinmann, G. Palmer, M. Emons, and U. Morgner, “Few-cycle OPCPA system at 143 kHz with more than 1 microJ of pulse energy,” Opt. Express 18(3), 2836–2841 (2010). [CrossRef] [PubMed]
13. C. Homann, C. Schriever, P. Baum, and E. Riedle, “Octave wide tuneable UV-pumped NOPA: pulses down to 20 fs at 0.5MHz repetition rate,” Opt. Express 16(8), 5746 (2008). [CrossRef]
14. J. Bromage, J. Rothhardt, S. Hädrich, C. Dorrer, C. Jocher, S. Demmler, J. Limpert, A. Tünnermann, and J. D. Zuegel, “Analysis and suppression of parasitic processes in noncollinear optical parametric amplifiers,” Opt. Express 19(18), 16797–16808 (2011). [CrossRef] [PubMed]
15. M. Bradler, P. Baum, and E. Riedle, “Femtosecond continuum generation in bulk laser host materials with sub-µJ pump pulses,” Appl. Phys. B 97(3), 561–574 (2009). [CrossRef]
16. G. Cerullo, M. Nisoli, and S. De Silvestri, “Generation of 11 fs pulses tunable across the visible by optical parametric amplification,” Appl. Phys. Lett. 71(25), 3616 (1997). [CrossRef]
17. A. M. Siddiqui, G. Cirmi, D. Brida, F. X. Kärtner, and G. Cerullo, “Generation of <7 fs pulses at 800 nm from a blue-pumped optical parametric amplifier at degeneracy,” Opt. Lett. 34(22), 3592–3594 (2009). [CrossRef] [PubMed]
18. R. Antipenkov, A. Varanavičius, A. Zaukevičius, and A. P. Piskarskas, “Femtosecond Yb:KGW MOPA driven broadband NOPA as a frontend for TW few-cycle pulse systems,” Opt. Express 19(4), 3519–3524 (2011). [CrossRef] [PubMed]
19. E. Riedle, M. Beutter, S. Lochbrunner, J. Piel, S. Schenkl, S. Spörlein, and W. Zinth, “Generation of 10 to 50 fs pulses tunable through all of the visible and the NIR,” Appl. Phys. B 71, 457–465 (2000).
20. T. Binhammer, E. Rittweger, R. Ell, F. X. Kärtner, and U. Morgner, “Prism-based pulse shaper for octave spanning spectra,” IEEE J. Sel. Top. Quantum Electron. 41(12), 1552–1557 (2005). [CrossRef]