We have built a prepulse-free, multi-terawatt, ultrashort pulse laser system, which combines both conventional laser amplification and optical parametric chirped pulse amplification (OPCPA) techniques. By employing an OPCPA system after the regenerative amplifier in a Ti:sapphire chirped pulse amplification laser chain, we have dramatically enhanced the prepulse contrast by 6 orders of magnitude. A prepulse contrast of better than 4.4 × 10-11 has been measured with a high energy broadband pulse of 24 mJ at 10 Hz repetition rate from the OPCPA system. Using a subsequent four-pass Ti:sapphire amplifier, we have achieved an amplified energy of 279 mJ and an ultrashort recompressed amplified pulse duration of 23.5 fs, corresponding to the peak powers for OPCPA and four-pass amplifier of 0.5 TW and 5.9 TW, respectively.
© 2006 Optical Society of America
Ti:sapphire chirped pulse amplification (CPA) laser technology has made possible the development of table-top laser systems supporting multi-terawatt power level [1,2]. This progress has opened the way to investigate and exploit a wide range of high field interaction studies at high focused intensities on the order of 1020 W/cm2 by using small-scale laboratory systems. The important role of prepulse contrast was shown in high-intensity laser matter interaction processes, for example, in the development of ultrafast X-ray sources [3–5]. Notably, in the case of cluster targets, the prepulse could destroy the cluster formation, causing the main laser pulse to interact with a low density plasma. The prepulse, appearing about ten nanoseconds before the main pulse, results from the previous round trip in the regenerative-amplifier (RA) of the laser chain. Here, we will use the definition of contrast as the intensity ratio between the prepulse and the main pulse. Recently, Fukuda et al. demonstrated that hot electrons produced from atomic clusters by improving the prepulse contrast shift the ion balance toward the higher charge states, which dramatically enhances the x-ray line yield and also the ion kinetic energy . In this experiment, a laser pulse with a peak focused intensity of 3 × 1018 W/cm2 and contrast ratio of 5 × 10-6 was used. In order to study this type of experiment at focused intensities greater than 1020 W/cm2, it is essential that the level of prepulse be more than 10 orders of magnitude below the main pulse to avoid the preplasma formation. The prepulse level is attenuated by using additional Pockels cells after the RA, however, the prepulse contrast that is achieved in this manner is currently about 7 orders of magnitude below the main pulse. Also, contrast enhancement by means of Pockels cell cascading is difficult to implement at higher energies due to limitations of size and damage threshold of the crystals.
Optical parametric amplification  has recently been identified as an attractive amplification technology, and its use in CPA systems is referred to as optical parametric chirped pulse amplification (OPCPA) [8,9]. OPCPA permits significant reduction in the level of prepulse compared with the conventional CPA systems since the amplification occurs only within the time window defined by the pump pulse. No amplification of the prepulse occurs in OPCPA if the pump pulse width is shorter than the characteristic delay between prepulse and main pulse. Another advantage of this technique is the broadband capability similar to that of Ti:sapphire amplifier, which can support ultrashort laser pulses. The short pulse duration is essential to realize compact, high peak power table-top CPA lasers. By using the OPCPA technique, a prepulse contrast level of 10-8 has been demonstrated recently [10,11]. However, this value is not high enough to avoid perturbations in the laser-matter interaction process at focused intensities greater than 1020 W/cm2. Although contrast levels exceeding 10 orders of magnitude can be accomplished by achieving a gain of 1010 in OPCPA, even though the seed pulses are attenuated to the pJ level, this still requires a several joule pump laser which is impractical for compact, low-cost table-top applications [12,13]. A more practical scheme would be one in which the seed pulses from the oscillator are first passed through the Pockels cell in order to enhance the prepulse contrast and then amplified via OPCPA. However, several OPCPA stages (typically more than five) would be required to produce multi-millijoule energies . Also, for the noncollinear OPCPA geometry, the amplified pulse spectrum is very sensitive to the pump-signal angle . Small variations of this angle would significantly reduce the amplified bandwidth as well as gain and results in a significant distortion of the amplified spectrum . This modulation in the spectrum has serious disadvantages for further amplification following pulse recompression and for long-term operation.
In this paper, we demonstrate a prepulse-free, multi-terawat, sub-30-fs laser system, realized by a simple and effective technique, so called REPA (REgenerative and Parametric Amplifiers), that combines the advantage of OPCPA with RA in a Ti:sapphire CPA laser chain. Pulses from RA with additional Pockels cells typically have a prepulse contrast of 6 orders magnitude with ~10 millijoule level energy. The pulse energy is attenuated to ~10 nanojoule level while maintaining the same prepulse contrast. The main pulse is then selectively amplified in OPCPA to ∼10 millijoule level again. This amplification by 6 orders of magnitude from ∼10 nanojoules to ∼10 millijoules can be readily obtained in a single pass through only a few cm of OPCPA pumped by a commercially available low-energy pump laser . Typically, the pump lasers have a pulse duration of 5-10 ns which is short in comparison to the time delay between the prepulse and the main pulse. The main pulse can therefore be selectively amplified in the OPCPA, leading to a high contrast. Since the contrast enhancement in an OPCPA is numerically equal to the amplification gain, with a gain of 6 orders of magnitude for example, the prepulse contrast will be 12 orders of magnitude after REPA. The pulse energy before the RA is typically sub-nanojoules or less. For amplifying this in the OPCPA directly without the RA, and to obtain same output energy, additional and larger sized OPCPA stages would be required . This would be impractical for further amplification following pulse recompression, and for long-term operation , as noted above. Since the efficiency of the RA is much higher compared to that of OPCPA, in spite of the fact that the RA output is attenuated, the total pump energy required for our REPA is lower compared to a system without RA. REPA can be easily configured by just adding a compact OPCPA system to the existing CPA systems in widespread use without any major change in the original system. With REPA, we have succeeded in producing an extremely high prepulse contrast of better than 4.4 × 10-11. REPA is also appropriate for use with high energy pulses (tens of millijoules or more). This would have a significant advantage for further amplification, since the subsequent amplifiers present low gain and therefore produce less prepulse amplification. By seeding high energy pulses into the subsequent amplifiers, one can expect intense, high contrast pulses at the output of the laser chain. In fact, we have produced high energy pulses of 24 mJ from the OPCPA in our REPA system. We have amplified this further to 279 mJ in the subsequent four-pass Ti:sapphire amplifier and have achieved a recompressed amplified pulse duration of 23.5 fs, corresponding to a peak power of 5.9 TW. To the best of our knowledge, it is the first hybrid laser system, which combines both conventional Ti:sapphire CPA and OPCPA techniques, to produce multi-terawatt peak powers with sub-30-fs pulses and the first to demonstrate prepulse contrast levels exceeding 10 orders of magnitude.
2. Experimental setup
Figure 1 shows the schematic of the prepulse-free, multi-terawatt laser system. The OPCPA system was employed after the RA in a part of our PW Ti:sapphire CPA laser chain [1,16]. A mode-locked Ti:sapphire oscillator produced 10 fs seed pulse at the center wavelength of 800 nm. Before amplification, the seed pulse was temporally expanded to 1 ns (FWHM) by the stretcher and propagated through a Faraday isolator that provides sufficient optical isolation of the oscillator from the amplifier section. The stretched seed pulse with 0.02 nJ energy was then amplified in RA to ~5 mJ with ~65 nm bandwidth (FWHM). Two Pockels cells were placed after the RA to enhance the prepulse contrast to 10-6 and then the seed pulse energy was attenuated to ~5 nJ while maintaining the contrast at the same level. The attenuated seed pulse was collimated and relayed to the OPCPA.
Figure 2 schematically illustrates the OPCPA system. The OPCPA was designed to operate with two amplification stages, since the maximum individual gain was limited by the optical parametric generation. We limited the highest gain of each OPCPA stage to 104 in order to avoid possible optical parametric oscillation caused by the small reflective losses of the antireflection-coated optical crystal surfaces. With this gain, we also expect the parametric fluorescence to be less than 1% . BBO crystal was chosen as the nonlinear crystal because of its high effective nonlinear coefficient. The first and second stage crystals had a cross-section of 7 mm × 7 mm and their lengths were 16 mm and 19.5 mm, respectively. Both antireflection-coated crystals were cut at 23.8° to maximize the gain bandwidth and to facilitate type I noncollinear angular phase matching with an external noncollinear angle between the seed and the pump (pump-signal angle) of 3.9°. Each crystal was mounted on a precision rotation stage to optimize the angle between the input beams and the crystal.
The pump pulse was generated by a commercial Q-switched single-longitudinal mode seeded Nd:YAG laser operating at 532 nm (Spectra-Physics GCR-150), which produces a 10 Hz pulse train of 300 mJ, 5.5 ns (FWHM) pulses. The output temporal profile was observed to be smooth and near-Gaussian, and the spatial profile almost flat-top. The output from pump laser was split using a combination of wave plates and polarizers to produce the pump pulses for the two OPCPA stages. The pump beams were relay imaged to each stage by means of two vacuum image telescopes, which simultaneously adjusted the pump beam diameter, in order to maintain a uniform intensity profile and prevent optics damage caused by Fresnel diffraction. The seed and pump pulses were synchronized in time, with a characteristic timing jitter of ± 0.76 ns in our system.
The pump beam diameter for each stage was 2.75 mm at 1/e2 intensity points and the maximum pump pulse energy was 145 mJ, which results in 2.4 J/cm2 fluence and 450 MW/cm2 pump intensity maximum for the 5.5 ns pump pulse. The diameter of the vertically polarized seed beam was 2.4 mm at 1/e2 intensity points in both stages. The pump pulse was horizontally polarized and was injected from behind the final seed steering mirrors. The center of the seed beam was displaced from the center of the pump beam by ~0.18 mm in the sensitive phase-matching direction on the front face of both crystals in order to maximize the spatial overlap of the beams for high pump-to-signal conversion efficiency.
3. Results and discussion
We measured the amplification gain of the OPCPA system by using a photodiode together with a set of calibrated neutral density filters. The gain values for each individual stage as a function of pump intensity are shown in Fig. 3. The energy of the input seed pulse was 4 nJ, corresponding to ~90 W/cm2 intensity. A maximum gain of ~3300 was achieved for the first stage with a pump intensity of 424 MW/cm2 [Fig. 3(a)]. A strong saturated gain of ~1900 was obtained at the second stage with 465 MW/cm2 pump intensity [Fig. 3(b)]. The OPCPA system amplified the 4 nJ seed pulse to as much as 24 mJ with 290 mJ total pump pulse energy, corresponding to a gain of over 6 × 106 and a pump-to-signal energy conversion efficiency of 8.3 %. The saturation of the second stage is desirable because it extracts a large amount of energy from the system and also leads to an increase in stability. The measured energy stability in the saturation condition was 1.25 % (one standard deviation of 18,000 consecutive shots).
Figure 4 shows the spectra of the amplified pulse obtained by a combination of regenerative pulse shaping  and OPCPA for different phase matching conditions. In OPCPA, the change of the phase matching condition makes it possible to tune the central wavelength of the amplified pulse and therefore rough spectral shaping of the amplified pulse in a given spectral range is possible by cascading two BBO crystals working under slightly different phase matching conditions. In addition, some spectral control can also be achieved by varying the non-collinear pump-signal angle and the pulse timing in different stages of amplification. However, it is not possible to obtain arbitrary, precisely shaped amplified pulses this way because amplification of only two central wavelengths can be controlled in OPCPA. The spectrum would be optimized better by regenerative pulse shaping to match the gain characteristics of the subsequent amplifiers. By regenerative pulse shaping and changing the phase matching condition in OPCPA, we confirmed and demonstrated that the amplified pulse was shaped spectrally. As can be seen from the spectra in Fig. 4, it is possible to obtain an arbitrarily shaped amplified pulse. The main parameter is the pump-signal angle a as indicated in the figure. A combination of regenerative pulse shaping and OPCPA is a very simple way to achieve arbitrary spectral pulse shaping and should permit the production of multi-terawatt power levels with ultrafast pulse duration. This is another significant advantage of our REPA system.
The spatial intensity profiles for the pump, input signal to the OPCPA and amplified output signal from the OPCPA are shown in Fig. 5. As seen from this figure, an excellent beam quality was obtained after amplification in the OPCPA showing that optical parametric amplification process conserves the signal beam quality to a high degree.
We measured the prepulse contrast from our REPA system by using a photodiode and a set of calibrated filters. For this, we used a high extinction ratio polarizer and Pockels cell combination to branch off the main pulse so as to measure the extremely small prepulse intensity level (<10-10 of main pulse), while avoiding damage of the photodiode. Additionally, we focused the whole beam into the photodiode by a lens through a long shielding tube, increasing the signal-to-noise ratio. Figure 6 shows the temporal contrast of the amplified pulse from RA [Fig. 6(a)] and REP A [Fig. 6(b)]. The prepulse contrast was measured to be 10-6 in Fig.6(a). However, no prepulse resulting from leakage of the RA was observed in Fig. 6(b) and therefore the prepulse contrast was better than 4.4 × 10-11 (detection limited). In Fig. 6 the main pulse is strongly saturated. The contrast of the incident seed pulse for the OPCPA system was 10-6, as shown in Fig. 6(a). In our case, the prepulse was located 10.9 ns before the main pulse, and the pump pulse duration was 5.5 ns (FWHM). Therefore, the pump intensity dropped to ~0.038 MW/cm2 at the position of the prepulse under the maximum pumping condition and the amplification of the prepulse in OPCPA system was calculated to be ~1.18. This figure is negligible compared to the main pulse amplification value of ~107. Therefore, in our OPCPA, the contrast enhancement is numerically equal to its gain, and the prepulse contrast obtained in our REPA experiment was estimated to be over 6 × 10-12. Although amplified-spontaneous-emission (ASE) may also affect preplasma formation, this can be removed via specific optical schemes such as a high-energy seed pulse injection into the RA , the use of a saturable absorber  or a nonlinear induced birefringence .
The output pulse from REPA was up-collimated to ~6 mm diameter and then introduced into the four-pass Ti:sapphire amplifier for further amplification. We used in this amplifier a 15 mm-long Ti:sapphire crystal pumped by 532 nm radiation at 10 Hz. The amplified pulse spectrum in the Ti: sapphire amplifier was red-shifted due to saturation. The spectrum of the input pulse for the amplifier was therefore shifted to shorter wavelength side by tuning the phase matching condition of the BBO crystals in OPCPA. Figure 7 shows the output energy from the four-pass amplifier as a function of pump energy. The maximum seed energy obtained from the four-pass amplifier was 279 mJ using a pump energy of 768 mJ. We carried out recompression by using a grating compressor optimized to give minimum recompressed pulse duration with 50 % efficiency. The typical autocorrelation trace and amplified spectrum after compressor are shown in Fig. 8. The measured FWHM pulse duration was 23.5 fs, corresponding to peak powers for the OPCPA and four-pass amplifier of 0.5 TW and 5.9 TW, respectively. In order to evaluate the compressibility of the amplified pulse, we calculated the duration of the transform limit from the measured amplified spectrum of over 70 nm (FWHM) after compression. The calculation predicts a recompressed pulse duration of 20 fs, which agrees reasonably well with the measured value.
We have produced prepulse-free, multi-terawatt and sub-30-fs pulses by REP A that employs the OPCPA stage after RA in the existing Ti:sapphire CPA laser chain. REPA exhibits superior prepulse contrast compared with systems based on Ti:sapphire alone. A prepulse contrast enhancement by several orders of magnitude and to better than 4.4 × 10-11 (detection limited) was demonstrated. This technique can readily be applied to current femtosecond Ti:sapphire CPA systems in widespread use. The 800 nm broadband chirped-pulse was amplified to 24 mJ in the OPCPA. We have also demonstrated that the amplified pulse can be shaped spectrally with REPA. We amplified this further to 279 mJ in the subsequent four-pass Ti:sapphire amplifier and achieved a recompressed amplified pulse duration of 23.5 fs. The contrast, wavelength and pulse energy level from the demonstrated REPA are ideal for seeding high-energy, large-Ti: sapphire-amplifiers of a PW laser chain. The idea of REPA can also be extended and utilized as a part of the front end of a multi-kilojoule short pulse Nd:glass laser operating at 1µm wavelength.
The authors thank Y. Nakai and Y. Yamamoto for technical contribution, and T. Harimoto and Y. Fukuda for many fruitful discussions.
References and links
1. K. Yamakawa, M. Aoyama, S. Matsuoka, T. Kase, Y. Akahane, and H. Takuma, “100-TW sub-20-fs Ti:sapphire laser system operating at a 10-Hz repetition rate,” Opt. Lett. 23, 1468–1470 (1998). [CrossRef]
2. M. Pittman, S. Ferré, J. P. Rousseau, L. Notebaert, J. P. Chambaret, and G. Chériaux, “Design and chracterization of a near-diffraction-limited femtosecond 100-TW 10-Hz high-intensity laser system,” Appl. Phys. B 74, 529–535 (2002). [CrossRef]
3. J. C. Kieffer, Z. Jiang, A. Ikhlef, and C. Y. Cote, “Picosecond dynamics of a hot solid-density plasma,” J. Opt. Soc. Am. B 13, 132–137 (1996). [CrossRef]
4. U. Andiel, K. Eidmann, and K. Witte, “Time-resolved x-ray K-shell spectra from high density plasmas generated by ultrashort laser pulses,” Phys. Rev. E 63, 026407-1–026407-7 (2001). [CrossRef]
5. T. Caillaud, F. Blasco, F. Dorchies, Y. Glinec, Ch. Stenz, and J. Stevefelt, “Experimental study of K-shell X-ray emission from argon clusters irradiated by an ultra-intense laser pulse,” Nucl. Instrum. Methods Phys. Res. B 205, 329–333 (2003). [CrossRef]
6. Y. Fukuda, Y. Akahane, M. Aoyama, N. Inoue, H. Ueda, Y. Kishimoto, K. Yamakawa, A. YA. Faenov, A. I. Magunov, T. A. Pikuz, I. Yu. Skobelev, J. Abdallah, Jr., G. Csanak, A. S. Boldarev, and V. A. Gasilov, “Generation of X rays and energetic ions from superintense laser irradiation of micro-sized Ar clusters,” Laser and Particle Beams 22, 215–220 (2004). [CrossRef]
7. R. A. Baumgartner and R. Byer, “Optical parametric amplification,” IEEE J. Quantum Electron. 15, 432–444 (1979). [CrossRef]
8. A. Dubietis, G. Jonusaukas, and Piskarskas, “Powerful femtosecond pulse generation by chirped and stretched pulse parametric amplification in BBO crystal,” Opt. Commun. 88, 437–440 (1992). [CrossRef]
9. I. N. Ross, P. Matousek, M. Towrie, A. J. Langley, and J. L Collier, “The prospect for ultrashort pulse duration and ultrahigh intensity using optical parametric chirped pulse amplifiers,” Opt. Commun. 144, 125–133 (1997). [CrossRef]
10. I. Jovanovic, C. A. Ebbers, and C. P. J. Barty, “Hybrid chirped-pulse amplification,” Opt. Lett. 27, 1622–1624 (2002). [CrossRef]
11. H. Yoshida, E. Ishii, R. Kodama, H. Fujita, Y. Kitagawa, Y. Izawa, and T. Yamanaka, “High-power high-contrast optical parametric chirped pulse amplification in P -BaB2O4 crystal,” Opt. Lett. 28, 257–259 (2003). [CrossRef]
12. I. N. Ross, J. L. Collier, P. Matousek, C. N. Danson, D. Neely, R. M. Allott, D. A. Pepler, C. Hernandez-Gomez, and K. Osvay, “Generation of terawatt pulses by use of optical parametric chirped pulse amplification,” Appl. Opt. 39, 2422–2427 (2000). [CrossRef]
13. X. Yang, Z. Xu, Y. Leng, H. Lu, L. Lin, Z. Zhang, R. Li, W. Zhang, D. Yin, and B. Tang “Multiterawatt laser system based on optical parametric chirped pulse amplification,” Opt. Lett. 271135–1137 (2002). [CrossRef]
14. I. Jovanovic, C. Brown, B. Wattellier, N. Nielsen, W. Molander, B. Stuart, D. Pennington, and C. P. J. Barty, “Precision short-pulse damage test station utilizing optical parametric chirped-pulse amplification,” Rev. Sci. Instrum. 75, 5193–5202 (2004). [CrossRef]
15. X. Yang, Z. Xu, Z. Zhang, Y. Leng, J. Peng, J. Wang, S. Jin, W. Zhang, and R. Li, “Dependence of spectrum on pump-signal angle in BBO-I noncollinear optical-parametric chirped-pulse amplification,” Appl. Phys. B 73, 219–222 (2001). [CrossRef]
17. C. P. J. Barty, G. Korn, F. Raksi, C. Rose-Petruck, J. Squier, A.-C. Tien, K. R. Wilson, V. V. Yakovlev, and K. Yamakawa, “Regenerative pulse shaping and amplification of ultrabroadband optical pulses,” Opt. Lett. 21, 219–221 (1996). [CrossRef]
18. J. Itatani, J. Faure, M. Nantel, G. Mourou, and S. Watanabe, “Suppression of the amplified spontaneous emission in chirped-pulse-amplification lasers by clean high-energy seed-pulse injection,” Opt. Commun. 148, 70–74 (1998). [CrossRef]
19. A. Jullien, F. A-Rochereau, G. Chériaux, J-P. Chambaret, P. d’Oliveira, T. Auguste, and F. Falcoz, “High-energy, simple setup for pulse cleaning at the millijoule level by nonlinear induced birefringence,” Opt. Lett. 29, 2184–2186 (2004). [CrossRef]