Femtosecond green pulses were generated from broadband pulses centered at 800 nm and quasi-monochromatic pulses centered at 532 nm using noncollinear optical parametric chirped pulse amplification (NOPCPA) followed by sum frequency mixing. In addition to amplifying the 800-nm pulses, the NOPCPA stage pumped by a Q-switched, injection seeded Nd:YAG laser also provided broadband idler pulses at 1590 nm. The signal and idler pulses were sum frequency mixed using achromatic and chirp assisted phase matching yielding pulses near 530 nm with a bandwidth of 12 nm and an energy in excess of 200 μJ. The generated pulses were recompressed with a grating compressor to a duration of 150 fs. The technique is scalable to high energies, broader bandwidths, and shorter pulse durations with compensation for higher order chirps and dedicated engineering of the interacting beams.
©2011 Optical Society of America
High-power ultrashort pulses in the visible spectral range have important applications in filamentation nonlinear optics and mixing, micro- and nano-machining, and the study of ultrafast processes in biological and molecular systems. The generation of such radiation is typically based on frequency upconversion of femtosecond, near-infrared pulses. Femtosecond non-collinear optical parametric amplifiers (NOPAs) seeded by white light continuum and pumped by the second harmonic of a Ti:sapphire laser constitute a popular fs source in the visible. Such blue-pumped NOPA systems exhibit an ultrabroad phase-matching bandwidth from 500 to 750 nm [1–3] and were recently shown to be scalable to a few hundred μJ . However, some of the applications above require further increase in pulse energy in certain ranges within the visible. For example, nanosphere assisted patterning, which can be used to produce bio-sensors , requires light sources tuned to the resonant wavelength of the patterning process. The resonance often falls into the visible spectral range  and large pulse energies are needed to reach the critical fluence for material modifications over an extended surface area.
With the development of ultrafast multi-petawatt optical amplifier systems, reaching ultra-high temporal contrast becomes imperative. With focused intensities above 1022 Wcm–2, the temporal contrast must exceed 1012 to avoid pre-ionization of solid targets even with high damage thresholds. Optical parametric (chirped pulse) amplification (OPCPA) [3, 7–9], where amplified fluorescence and amplification of any pre- or post-pulses are confined to the duration of the pump pulse, may offer improved contrast compared to amplifiers based on laser active media, especially if deep saturation is avoided . Often, one needs further temporal filtering to reach the required contrast. In addition to saturable absorption , polarization ellipse rotation (PER) , cross polarized wave generation (XPW) [13–15], and plasma mirrors , frequency upconversion has also been considered as a possibility [17–19]. One attempt towards a high power laser system is the use of the X-A* transition of XeF in an excimer laser  seeded by a frequency doubled OPCPA chain. This approach is very promising, provides high temporal contrast, but is limited to the wavelength of 475 nm .
In this paper, we introduce a feasibility study and a pilot experiment for efficient conversion of femtosecond 800-nm pulses of a common (multi)-TW chirped pulse amplification (CPA) laser system based on Ti:sapphire into broadband high-power pulses in the green spectral range. The frequency conversion is performed in two stages at stretched pulse durations, which in principle allows high conversion efficiency without introducing nonlinear pulse distortions even in crystals with limited aperture size. This is in contrast to the fs, blue-pumped NOPA, where the requirement of large-aperture, thin crystals due to the high pulse intensities render energy scaling problematic and expensive. Due to the two consecutive frequency conversion steps of OPCPA and sum frequency generation (SFG), the technique can potentially offer ultrahigh temporal contrast as well.
2. Basic scheme
The basic scheme for the generation of powerful light pulses in the visible is depicted in Fig. 1. An OPCPA in non-collinear configuration amplifies broadband, chirped seed pulses at 800 nm. As an inherent feature, energetic broadband idler pulses are also generated, which in typical systems have been considered mostly as a side effect and dumped. Now the signal and idler pulses are sum frequency mixed, where the extension of the phase matching bandwidth is achieved by the simultaneous application of the techniques of achromatic phase matching (APM) [21–24] and chirp assisted phase matching (CPM) . The two consecutive processes, parametric amplification and sum frequency mixing, can be considered as converting the high-energy, quasi-monochromatic pump pulses of a frequency doubled Nd:YAG laser into energetic, broadband pulses around its center wavelength.
3. Theoretical analysis
Energy conservation dictates that the angular frequencies in a three-wave mixing process satisfy the equation ω p(t) = ω s(t) + ω i(t), where the subscripts p, s, and i refer to pump, signal, and idler, respectively. Assuming linear temporal chirp, ω(t) = ω 0 +βt, where β is the linear chirp parameter, one can show that the chirp parameters satisfy the equation β p = β s +β i.
As our OPCPA was pumped by quasi-monochromatic pulses, the temporal chirp in the generated idler wave was the inverse of that of the amplified seed,Fig. 2(a). Mixing the direct output beams of the OPCPA would lead to a quasi-monochromatic sum frequency wave, i.e. the inverse process of parametric amplification. In order to avoid that, the chirp and angular dispersion of the interacting beams have to be changed appropriately.
In general, broadband SFG can be achieved using noncollinear geometry, temporal chirp manipulation, and angularly dispersed beams. As we require no angular dispersion in the sum frequency wave, both the signal and idler waves have to be angularly dispersed as shown in Fig. 2(b).
Temporal chirp manipulation or CPM can lead to an enhancement of the bandwidth of frequency conversion by overlapping proper pairs of spectral components from each of the broadband, chirped input beams. In addition to applications to SFG and DFG [25, 26], CPM was also suggested and then applied to OPCPA [7, 27, 28]. By modifying the temporal chirp of one or both output beams from the OPCPA prior to the SFG process, a finite bandwidth can be generated in the sum frequency wave andEq. (3) and the formula . The parameter that controls the bandwidth is the ratio of the idler and signal chirp parameter, . Figure 3 shows the bandwidth generated in the sum frequency mixing process relative to that of the signal as a function of the chirp ratio. The largest bandwidth is reached at with a value of leading to 2 times shorter transform limited sum frequency pulse duration compared to that of the signal pulse. In contrast, the bandwidth becomes extremely narrow at . The latter case corresponds to the situation of direct sum frequency mixing of the output of the OPCPA, resulting in the quasi-monochromatic bandwidth of the nanosecond pump field. At chirp ratios of , the central frequency of the generated pulse can be tuned within the maximum bandwidth region of by varying the relative delay between signal and idler pulses [26, 27].
Our goal in the feasibility study was the demonstration of the generation of high energy, broadband pulses in the visible with the use of our existing laser system  with the necessary modifications and a setup as simple as possible. A compromise among conversion efficiency, bandwidth, laboratory space and simple setup was found at a chirp ratio of . For seed pulses with a FWHM of 40 nm centered at 800 nm, this chirp ratio leads to a FWHM of 12 nm at 532 nm supporting a Fourier limit of ~35 fs in positively chirped pulses, which in principle can be compressed using a simple grating compressor (cf. dashed lines in Fig. 3). In addition, this chirp ratio can be produced by moderate compression of the positively chirped, amplified signal pulses in a grating compressor prior to SFG. We note that at a signal bandwidth and transform limited duration of 40 nm and 23.5 fs, respectively, and , the corresponding parameters of the sum frequency pulse would be 25 nm and 16.6 fs.
Achromatic phase matching is a term generally applied to describe nonlinear mixing, where each individual spectral component of the beam is propagating in the nonlinear medium at its own phase matching angle with the goal of achieving momentum conservation in an extended spectral range. This typically results in the use of angularly dispersed beams, i.e. pulse front tilt. The idler pulses generated in our OPCPA have an inherent angular dispersion of ~80 μrad/nm in the crystal. By requiring that the magnitude of the angular dispersion of the idler be unchanged prior to SFG in a type-I β-BaB2O4 (BBO), we can determine the angular dispersion of the signal pulses that leads to phase matching at a chirp ratio of . The results are shown in Fig. 4 by the dashed lines. Broadband phase matching is achieved at θ = 27° and the required angular dispersion in the signal wave is 97 μrad/nm in the crystal (160 μrad/nm in air). The corresponding angle between the signal and the pump beam in the crystal is 4.1°, while the angle between the signal and idler beam is 12°. It is also important to note that all three pulse fronts overlap approximately under these conditions and the group velocities are matched, which leads to optimum conversion efficiency.
4. Experimental scheme and results
Figure 5 shows the schematic of the pilot experiment. The pump source was a 10-Hz, Q-switched, single longitudinal mode Nd:YAG laser (Spectra Physics, Quanta Ray) delivering 400 mJ at 532 nm with a pulse duration of 7.5 ns at FWHM. The pump was split into two main beams. One beam with a pulse energy of 160 mJ was used to pump a high contrast, hybrid amplifier , which provided 9-mJ, positively chirped pulses with a duration of 150 ps and a bandwidth of 40 nm. The second, 240-mJ Nd:YAG beam was spatially filtered and relay imaged onto a noncollinear OPCPA using a delay line of 17.5 m.
The output of the laser system was split into two beams. One beam with a pulse energy of 2.7 mJ was used as the seed for the OPCPA and to generate broadband idler pulses at 1.6-μm. The OPCPA was based on a 7-mm-long type I BBO, AR coated on both sides, cut at a phase matching angle of 23.8°. The input diameter of the pump and the seed pulses were 5 mm and 3.5 mm, respectively. Due to the modest temporal overlap between the input beams (i.e. only 8 mJ of pump energy was available in a 300-ps window), the generated idler pulse energy was limited to 600 μJ. The average incident pump intensity on the BBO had to be kept below 110 MW/cm2 to avoid damage on the coating, and therefore 25% of the pump pulse energy was converted into signal and idler photons. The amplified seed was blocked after the OPCPA.
Regarding the implementation of achromatic sum frequency generation, the idler pulses were then sent through a 1:1 telescope, which imaged the OPCPA crystal onto the SFG crystal. Imaging is important for avoiding spatial chirp and extra temporal chirp. The energy of the idler pulses incident on the SFG stage was 450 μJ. The second beam of the laser system with a pulse energy of 5.4 mJ was sent through a delay line, where the temporal chirp and angular dispersion required for broadband SFG were introduced into the pulses. The chosen idler-signal chirp ratio of −0.3 was produced by sending the signal beam through a grating compressor that reduced its pulse duration by a factor of 3.3. A 45°-apex-angle BK7 prism and a 4:1 telescope were used to generate the required angular dispersion into the beam and to image the prism surface onto the SFG crystal with a beam diameter of 3 mm. Due to losses in the compressor and the prism surfaces, only 1.5 mJ of signal was incident on the SFG stage. The fine adjustment of the angle between the signal and idler beams, and the phase matching angle of the SFG were accomplished by monitoring the residual angular dispersion in the sum frequency wave using spectrally and spatially resolved interferometry . Figure 6 shows the fringes obtained with a poorly (a) and a well aligned (b) SFG stage with wavefront angular dispersion values of 45.9 ± 0.7 μrad/nm and 1.5 ± 0.2 μrad/nm, respectively. Under our experimental conditions, the latter value leads to negligible lengthening. The generated sum frequency pulses had an energy of 210 μJ corresponding to a total energy conversion efficiency of 11%. Figure 7 shows the generated spectrum with a FWHM of 12 nm, which agrees well with the prediction. We note that an incorrect noncollinear angle, phase matching angle or sign of angular dispersion lead to similar sum frequency pulse energies and spectral bandwidth, albeit with an angularly dispersed output. The fact that the center wavelength in Fig. 7 is 527 nm and not 532 nm is a result of the particular time delay between the signal and idler pulses used in the experiment.
The positively chirped sum frequency pulses were recompressed using a compressor based on 1800 line/mm transmission gratings with 94% diffraction efficiency at 532 nm (Wasatch Photonics). The shortest pulses were obtained at a perpendicular grating distance of 27 cm, which compares well with the predicted value of 24 cm. Figure 8 shows the measured second order autocorrelation using a 50-μm-thick BBO. The FWHM of the autocorrelation trace was 200 fs, which corresponds to a pulse duration of approximately 150 fs, 4 times longer than the bandwidth limit calculated from the Fourier transform of the spectrum. The broad wings in the autocorrelation trace were tentatively attributed to higher-order chirp due to (i) higher-order chirp in the signal and idler pulses, (ii) higher-order angular dispersion in the signal beam prior to SFG, and (iii) residual high-order chirp due to the mismatch between the groove densities of the compression grating of the green pulses (1800 line/mm) and the 1200-line/mm diffraction gratings of the stretcher and compressor of the host laser system. By varying the angle of incidence on the transmission grating, in principle it is possible to tune the ratio of group delay dispersion and third-order dispersion and further minimize the pulse duration. However, our transmission gratings had high efficiency only in a narrow range of angles of incidence. We note that it is possible that effect (i) has a significant influence on the compressibility. Nevertheless, the study of high order chirp transfer in the sum frequency mixing process is beyond the scope of this paper. In general, compensation of higher-order chirp is possible by using a proper combination of gratings and prisms, or simply by using an acousto-optical programmable modulator.
With the use of picosecond pump pulses, the scheme may provide high intensity laser pulses with outranging temporal contrast. The two consecutive frequency conversion steps are expected to yield an output contrast that is larger than or equal to the square of the input contrast. Typically, the signal and idler pulses after an OPCPA stage exhibit a contrast that is at least as high as that of the input. The temporal profile of the sum frequency pulse is proportional to the product of the time dependent intensities of the two generating pulses and can lead to a quadratic scaling of the contrast under ideal conditions. Even under depletion, the contrast after SFG is typically not reduced by more than an order of magnitude compared to the undepleted case. In comparison, schemes based on saturable absorption , PER , and XPW  lead to only a constant, ~4 orders of magnitude improvement in contrast. We note that as XPW and PLR are third-order processes, they in principle lead to a higher contrast than second-order SFG. However, the contrast in those cases is limited by the extinction ratio of the polarizer-analyzer pair used in practice.
The scheme also has the potential to reach high energies. Scaling with respect to pulse energy is possible with (i) matching the pump and seed pulse durations in the OPCPA stage, (ii) increasing the aperture size of the crystals and the pump energy, and (iii) further amplification of the idler pulses using the fundamental of the Q-switched pump laser.
In addition, the technique can be implemented at other pump wavelengths as well, where broadband amplifying materials are available to compensate for the losses due to the frequency conversion process. Using high-energy, narrow-band pump sources in the near-infrared based on the fundamental wavelength of, for example, (i) Cr-doped materials such as Cr:LiSAF, Cr:LiCAF, (ii) Nd-doped YAG and glasses, or (iii) Yb-doped YAG, fluorides, and glasses, the sum frequency pulses would be generated in the primary spectral region for the current ultra-high peak power system development.
We have introduced a new procedure for converting ultrashort light pulses to the visible. The scheme was based on efficient broadband sum frequency mixing between the signal and idler output of a high-power, noncollinear optical parametric chirped pulse amplifier. The generated pulses could be compressed by a grating compressor to a duration approximately 4 times longer than the bandwidth limit. Further compression is possible with high-order chirp compensation either before or after sum frequency generation. With the use of the unique combination of well synchronised ultrashort pulses in the visible and in the infrared, the first successful experiment has also been implemented on the demonstration of an ultrafast optical switch based on a protein . The characterization of the beam quality and temporal contrast will be the subject of a future study. With straightforward modifications, the technique has the potential to generate high-energy, wide-bandwidth, ultrahigh contrast pulses in the near-infrared.
This work was jointly supported by the Hungarian Scientific Research Found (OTKA) under grant No K75149 and the National Office for Research and Technology Developments (NKTH) under grant No OTKA-NKTH CNK 78459. The research leading to these results has received funding also from the EC’s Seventh Framework Programme ( LASERLAB-EUROPE, grant agreement n° 228334).
References and links
1. T. Wilhelm, J. Piel, and E. Riedle, “Sub-20-fs pulses tunable across the visible from a blue-pumped single-pass noncollinear parametric converter,” Opt. Lett. 22, 1494–1496 (1997). [CrossRef]
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, 2268–2270 (1999). [CrossRef]
3. G. Cerullo and S. De Silvestri, “Ultrafast optical parametric amplifiers,” Rev. Sci. Instrum . 74, 1–18 (2003). [CrossRef]
4. P. Tzankov, J. Zheng, M. Mero, D. Polli, C. Manzoni, and G. Cerullo, “300 μJ noncollinear optical parametric amplifier in the visible at 1 kHz repetition rate,” Opt. Lett. 31, 3629–3631 (2006). [CrossRef] [PubMed]
5. A. Sipos, H. Tohati, A. Szalai, A. Mathesz, M. Gorbe, T. Szabo, M. Szekeres, B. Hopp, M. Csete, and I. Dekany, “Plasmonic structure generation by laser illumination of silica colloid spheres deposited onto prepatterned polymer-bimetal films,” Appl. Surf. Sci. 255, 5138–5145 (2009). [CrossRef]
7. I. N. Ross, P. Matousek, G. H. C. New, and K. Osvay, “Analysis and optimization of optical parametric chirped pulse amplification,” J. Opt. Soc. Am. B 19, 2945–2956 (2002). [CrossRef]
8. C. N. Danson, P. A. Brummitt, R. J. Clarke, J. L. Collier, B. Fell, A. J. Frackiewicz, S. Hawkes, C. Hernandez-Gomez, P. Holigan, M. H. R. Hutchinson, A. Kidd, W. J. Lester, I. O. Musgrave, D. Neely, D. R. Neville, P. A. Norreys, D. A. Pepler, C. J. Reason, W. Shaikh, T. B. Winstone, R. W. W. Wyatt, and B. E. Wyborn, “Vulcan petawatt: design, operation and interactions at 5 × 1020 Wcm–2,” Laser Part. Beams 23, 87–93 (2005). [CrossRef]
9. A. Dubietis, R. Butkus, and A. Piskarkas, “Trends in chirped pulse optical parametric amplification,” IEEE J. Sel. Top. Quant. Electron . 12, 163–172 (2006). [CrossRef]
10. F. Tavella, K. Schmid, N. Ishii, A. Marcinkevicius, L. Veisz, and F. Krausz, “High-dynamic range pulse-contrast measurements of a broadband optical parametric chirped-pulse amplifier,” Appl. Phys. B 81, 753–756 (2005). [CrossRef]
11. H. Kiriyama, M. Mori, Y. Nakai, T. Shimomura, M. Tanoue, A. Akutsu, H. Okada, T. Motomura, S. Kondo, S. Kanazawa, A. Sagisaka, J. Ma, I. Daito, H. Kotaki, H. Daido, S. Bulanov, T. Kimura, and T. Tajima, “Generation of high-contrast and high-intensity laser pulses using an OPCPA preamplifier in a double CPA, Ti:sapphire laser system,” Opt. Commun . 282, 625–628 (2009). [CrossRef]
12. 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]
13. A. Jullien, O. Albert, F. Burgy, G. Hamoniaux, J.-P. Chambaret, F. Auge-Rocherau, G. Cheriaux, J. Etchepare, N. Minkovski, and S. Saltiel, “10−10 temporal contrast for femtosecond ultraintense lasers by cross-polarized wave generation,” Opt. Lett. 30, 920–922 (2005). [CrossRef] [PubMed]
15. A. Jullien, C. G. Durfee, A. Trisorio, L. Canova, J.-P. Rousseau, B. Mercier, L. Antonucci, G. Cheriaux, O. Albert, and R. Lopez-Martens, “Nonlinear spectral cleaning of few-cycle pulses via cross-polarized wave (XPW) generation,” Appl. Phys. B 96, 293–299 (2009). [CrossRef]
16. G. Doumy, F. Quere, O. Gobert, M. Perdrix, P. Martin, P. Audebert, J. C. Gauthier, J. P. Geindre, and T. Wittmann, “Complete characterization of a plasma mirror for the production of high-contrast ultraintense laser pulses,” Phys. Rev. E 69, 026402 (2004). [CrossRef]
17. I. A. Begishev, M. Kalashnikov, V. Karpov, P. Nickles, H. Schonnagel, I. A. Kulagin, and T. Usmanov, “Limitation of second-harmonic generation of femtosecond Ti:sapphire laser pulses,” J. Opt. Soc. Am. B 21, 318–322 (2004). [CrossRef]
18. R. Clady, G. Coustillier, M. Gastaud, M. Sentis, P. Spiga, V. Tcheremiskine, O. Uteza, L. D. Mikheev, V. Mislavskii, J. P. Chambaret, and G. Cheriaux, “Architecture of a blue high contrast multiterawatt ultrashort laser,” Appl. Phys. B 82, 347–358 (2006). [CrossRef]
19. S. Y. Mironov, V. Lozhkarev, V. Ginzburg, I. V. Yakovlev, G. A. Luchinin, E. A. Khazanov, A. M. Sergeev, and G. Mourou, “Temporal intensity contrast ratio enhancement of petawatt level laser pulses based on second harmonic generation,” Proc. SPIE 7721, 77211R (2010). [CrossRef]
20. T. Hofmann, T. E. Sharp, C. B. Dane, P. J. Wisoff, W. L. Wilson Jr, F. K. Tittel, and G. Szabo, “Characterization of an ultrahigh peak power XeF(C → A) excimer laser system,” IEEE J. Quantum Electron . 28, 1366 (1992). [CrossRef]
21. V. D. Volosov and E. V. Goryachkina, “Compensation of phase-matching dispersion in generation of non-monochromatic radiation harmonics. I. Doubling of neodymium-glass radiation frequency under free-oscillation conditions,” Sov. J. Quantum Electron . 6, 854–857 (1976). [CrossRef]
22. G. Szabo and Z. Bor, “Broadband frequency doubler for femtosecond pulses,” Appl. Phys. B 50, 51–54 (1990). [CrossRef]
23. Y. Nabekawa and K. Midorikawa, “Group-delay-dispersion-matched sum-frequency mixing for the indirect phase control of deep ultraviolet pulses in the sub-20-fs regime,” Appl. Phys. B 78, 569–581 (2004). [CrossRef]
24. G. Arisholm, J. Biegert, P. Schlup, C. P. Hauri, and U. Keller, “Ultra-broadband chirped-pulse optical parametric amplifier with angularly dispersed beams,” Opt. Express 12, 518–530 (2004). [CrossRef] [PubMed]
25. K. Osvay and I. N. Ross, “Broadband sum-frequency generation by chirp-assisted group-velocity matching,” J. Opt. Soc. Am. B 13, 1431–1438 (1996). [CrossRef]
26. K. Osvay and I. N. Ross, “Efficient tuneable bandwidth frequency mixing using chirped pulses,” Opt. Commun . 166, 113–119 (1999). [CrossRef]
27. G. Veitas and R. Danielius, “Generation of narrow-bandwidth tunable picosecond pulses by difference-frequency mixing of stretched pulses,” J. Opt. Soc. Am. B 16, 1561–1565 (1999). [CrossRef]
28. Y. Tang, I. N. Ross, C. Hernandez-Gomez, G. H. C. New, I. Musgrave, O. V. Chekhlov, P. Matousek, and J. L. Collier, “Optical parametric chirped-pulse amplification source suitable for seeding high-energy systems,” Opt. Lett. 33, 2386–2388 (2008). [CrossRef] [PubMed]
29. K. Osvay, A. P. Kovacs, Zs. Heiner, M Csatari, Zs. Bor, G. Kurdi, M. Gorbe, J. Klebniczki, and I. E. Ferincz, “A table-top high contrast TW laser system,” Tech. Dig. CLEO Europe, CG-13-TUE (2005).
30. K. Varju, A. P. Kovacs, G. Kurdi, and K. Osvay, “High-precision measurement of angular dispersion in a CPA laser,” Appl. Phys. B 74, S259–S263 (2002). [CrossRef]
31. L. Fabian, M. Mero, Zs. Heiner, M. Kiss, K. Osvay, and A. Der, “Ultrafast integrated optical switching based on the Protein Bacteriorhodosin,” Tech. Dig. CLEO/QELS, CTuR7 (2010).