August 1, 2007 / Vol. 32, No. 15 / OPTICS LETTERS
2143
High-contrast optical-parametric amplifier as a front end of high-power laser systems C. Dorrer,* I. A. Begishev, A. V. Okishev, and J. D. Zuegel Laboratory for Laser Energetics, University of Rochester, 250 East River Road, Rochester, New York 14623, USA *Corresponding author:
[email protected] Received April 30, 2007; revised June 8, 2007; accepted June 8, 2007; posted June 11, 2007 (Doc. ID 82604); published July 23, 2007 A high-contrast preamplifier based on optical-parametric amplification with a short pump pulse is demonstrated. A gain larger than 105 and measurement-limited contrast higher than 1011 are obtained over a large temporal range extending within less than 10 ps of the peak of the pulse, because of the high instantaneous parametric gain provided by a short pump pulse in a nonlinear crystal. The energy gain and high contrast of this preamplifier make it a good seed source for high-power laser systems. © 2007 Optical Society of America OCIS codes: 140.4480, 320.7160.
The development of high-power laser systems has enabled the study of new regimes of laser–matter interactions [1]. Laser facilities, either demonstrated or under construction, have the potential for on-target intensities exceeding 1020 W / cm2 [2]. At these peak intensities, a degradation of the prepulse contrast, defined as the ratio of the peak power of the pulse to the peak power in a given temporal range before the main pulse, is detrimental to the interaction with the target since light absorbed before the main pulse can modify the target. The typical contrast reported for short-pulse laser systems is 108, approximately 100 ps before the main pulse, because of the laser fluorescence generated in chirped-pulse–amplification systems or parametric fluorescence generated in optical-parametric chirped-pulse-amplification systems (OPCPA) [3–5]. Other sources of contrast degradation include the short-pulse oscillator [6], leakage from regenerative amplifiers [4], and OPCPA pump-induced noise [7]. The contrast can be enhanced by nonlinear pulse-cleaning techniques [8,9], but reaching the energy levels required for efficient contrast enhancement can be complex. These techniques only have efficiencies of the order of 10%, and the contrast improvement is limited. OPCPA systems have demonstrated high extraction efficiency and bandwidth, but their contrast is dominated by parametric fluorescence extending over the temporal support of the pump [10]. For an optical parametric amplifier (OPA) in the nondepleted pump regime, the fluorescence intensity is proportional to the signal gain, which is, in turn, proportional to exp共␣冑Ipump兲, where Ipump is the pump intensity and ␣ is a constant characterizing the amplifier. The parametric fluorescence generated by a short pump pulse lies in a short temporal range, in contrast to laser fluorescence that extends over the lifetime of the corresponding laser transition. As the fluorescence intensity of both types of amplifiers is proportional to their gain [11], achieving part of the system gain in a preamplifier with no long-range fluorescence leads to a system contrast improvement approximately equal 0146-9592/07/152143-3/$15.00
to the preamplifier gain. We demonstrate an OPA with a gain higher than 105 and measurementlimited contrast better than 1011. A 1053 nm oscillator generates the OPA signal and pump for low jitter. The short 526.5 nm pump pulse provides high gain, and gates the input oscillator pulse and the parametric fluorescence to a few picoseconds. The front end is shown in Fig. 1. The 38 MHz oscillator provides 200 fs pulses centered at 1053 nm for the OPA signal after selection by an acousto-optic modulator and for the seed of the OPA pump, which is picked off and selected by a Pockels cell. The seed is stretched in a 75 m spool of single-mode fiber to avoid nonlinear effects during amplification. The seed is first amplified in a Nd:YLF diode-pumped regenerative amplifier (DPRA), where a nanosecond pulse synchronized with the short seed is injected in the DPRA to stabilize the amplification process, since the amplifier was originally designed for nanosecondpulse amplification. It was later found that the DPRA could be optimized for picosecond pulse amplification without the fiber spool and nanosecond pulse. After amplification and gain-narrowing, the duration of the pump at 1053 nm is approximately 8 ps, inferred from its autocorrelation, which is Gaussian with a FWHM of 12 ps. The DPRA is followed by a two-pass flash-lamp-pumped Nd:YLF head. Spatial apodiza-
Fig. 1. Diagram of the OPA based on a 38 MHz modelocked oscillator providing the OPA pump at 526.5 nm and signal at 1053 nm. © 2007 Optical Society of America
2144
OPTICS LETTERS / Vol. 32, No. 15 / August 1, 2007
tion of the Gaussian beam from the DPRA is performed with a round serrated-tooth apodizer to match the rod size in the Nd:YLF head and provide good conversion efficiency in the sum-harmonicgeneration and parametric process. The pulse amplified in the Nd:YLF head is frequency doubled in a 2.75 mm –barium borate (BBO) crystal, with energies up to 2 mJ used to pump the OPA. The pump at 526.5 nm is expected to be slightly shorter than the pulse at 1053 nm, considering the doubling conversion efficiency at ⬃50%. The pump energy in the OPA crystal [25 mm lithium triborate (LBO)] is varied by a wave plate and polarizer located before the dichroic mirror used to combine the pump and signal in a quasi-collinear geometry. The 2 mJ spatially superGaussian (⬃3 mm FWHM) temporally Gaussian (assumed ⬃6 ps FWHM) pump has an intensity of the order of 5 GW/ cm2. The synchronization between the signal at 1053 nm and the pump at 526.5 nm is achieved with an optical delay line. No jitter is observed because of the common seed for the pump and signal of the OPA, as reported previously for an OPCPA system [12]. The OPA repetition rate is 5 Hz determined by the repetition rate of the DPRA. The signal energy increases quasi-linearly with the pump energy and reaches 100 J for a 2 mJ pump, which corresponds to a signal gain of 1.5⫻ 105 (Fig. 2). The energy fluctuation of the amplified signal is correlated to the energy fluctuation of the DPRA and has an rms variation of 5% at 100 J. The unseeded OPA provides an upper bound of the fluorescence energy. The fluorescence energy is not measurable for signal energies below 40 J but rises linearly above this threshold. For these measurements, the amplified signal beam is free-propagating from the OPA crystal to the calorimeter, without spatial filtering. Figure 3 shows the spectral and temporal characteristics of the OPA. Spectral narrowing occurs at high conversion efficiency, and a temporal broadening correlated to the measured spectral narrowing is observed, although the spectral support of the pulses at the highest pump level remains 16 nm, which is sufficient for large-scale Nd:glass-based laser systems. This is attributed to depletion effects coupled with
Fig. 2. Energy of the signal (continuous curve) and energy of the fluorescence measured in the absence of seed in the OPA (dashed curve). The grayed region represents an interval of two standard deviations on the energy of the signal measured over 100 shots.
Fig. 3. Bandwidth of the signal defined as the FWHM of the spectrum (solid curve). FWHM of the measured autocorrelation (long dashed curve), and FWHM of the autocorrelation calculated with the FWHM of the measured spectrum (short dashed curve).
the temporal walk-off in the OPA crystal (approximately 1.2 ps for a 25 mm LBO), since the pump and signal intensities are comparable at high pump power. Self-phase modulation is not significant at intensities below 5 / GW cm2 in 2.5 cm of LBO, considering the measured nonlinear index of LBO, equal to 2.8⫻ 10−16 cm2 / W [13]. The second-order intensity autocorrelation of the unseeded OPA fluorescence for a 2 mJ pump has a FWHM of 3 ps and is well-fitted by a Gaussian function (Fig. 4). Using a decorrelation factor equal to 冑2, the FWHM of the fluorescence intensity is approximately 2.1 ps. For a Gaussian intensity profile, the fluorescence intensity is expected to decrease by 30 and 120 dB, 3.3 and 6.6 ps in front of its peak, respectively. This sets the limit for the achievable contrast, assuming that the parametric fluorescence is centered on the amplified pulse. The contrast of the amplified pulse was measured using a third-order scanning cross-correlator (Sequoia, Amplitude Technologies), where a fraction of the input pulse is doubled in a nonlinear crystal and used as a gate on the input pulse via sum-frequency generation in a second nonlinear crystal. Figure 5 displays the third-order cross-correlation of the amplified pulse at an energy of 50 J. No prepulse or pedestal is observed in front of the pulse down to 11
Fig. 4. Autocorrelation of the parametric fluorescence of the unseeded OPA for a 2 mJ pump on a linear (left) and logarithmic scale (right). The experimental data are plotted with a solid curve and the Gaussian fit is plotted with round markers.
August 1, 2007 / Vol. 32, No. 15 / OPTICS LETTERS
2145
In conclusion, we have demonstrated that an OPA pumped by a short-pump pulse can directly generate pulses with a measurement-limited temporal contrast better than 1011. The incoherent parametric fluorescence is temporally restricted around the pulse. Optical parametric amplifiers pumped by a short optical pulse are good candidates as highcontrast front ends for large-scale laser systems.
Fig. 5. Third-order cross-correlation of the amplified signal on a logarithmic scale.
This work was supported by the U.S. Department of Energy Office of Inertial Confinement Fusion under Cooperative Agreement DE-FC52-92SF19460, the University of Rochester, and the New York State Energy Research and Development Authority. The support of the DOE does not constitute an endorsement by the DOE of the views expressed in this Letter. The authors acknowledge the experimental assistance of Jake Bromage, Jason R. Brown, Moshe Frankel, and John M. Marciante from the Laboratory for Laser Energetics. References
Fig. 6. Detail of the third-order cross-correlation plotted in Fig. 5, where the pump is delayed from the signal by approximately 1 ps (solid curve) and of the third crosscorrelation measured when the pump is advanced by approximately 2 ps (dashed curve). The grayed region corresponds to the fluorescence calculated with a Gaussian intensity with a FWHM of 2 ps for comparison to the solid curve data.
orders of magnitude below the peak of the pulse, which is the diagnostic dynamic range for 100 J 300 fs pulses. Postpulses, which are not detrimental to most applications of high-power laser systems, are due to multiple reflections in the amplifier or crosscorrelator. The close-up of the measured crosscorrelation (Fig. 6) indicates that the contrast is preserved as close as 5 ps in front of the main pulse, although a prepulse at −7 ps was intermittently detected. This prepulse is attributed to a prepulse on the pump pulse. The fluorescence level is approximately −40 dB, but decreases rapidly to −110 dB at 8 ps in front of the pulse. Optimal contrast was observed when the pump was slightly delayed, with respect to the signal.
1. D. Umstadter, Phys. Plasmas 8, 1774 (2001). 2. J. D. Zuegel, S. Borneis, C. Barty, B. LeGarrec, C. Danson, N. Miyanaga, P. K. Rambo, C. LeBlanc, T. J. Kessler, A. W. Schmid, L. J. Waxer, J. H. Kelly, B. Kruschwitz, R. Jungquist, E. Moses, J. Britten, I. Jovanovic, J. Dawson, and N. Blanchot, Fusion Sci. Technol. 49, 453 (2006). 3. M. Nantel, J. Itatani, A.-C. Tien, J. Faure, D. Kaplan, M. Bouvier, T. Buma, P. Van Rompay, J. A. Nees, P. P. Pronko, D. Umstadter, and G. A. Mourou, IEEE J. Sel. Top. Quantum Electron. 4, 449 (1998). 4. H. Yoshida, E. Ishii, R. Kodama, H. Fujita, Y. Kitagawa, Y. Izawa, and T. Yamanaka, Opt. Lett. 28, 257 (2003). 5. V. Bagnoud, J. D. Zuegel, N. Forget, and C. Le Blanc, Opt. Express 15, 5504 (2007). 6. A. Braun, J. V. Rudd, H. Cheng, G. Mourou, D. Kopf, I. D. Jung, K. J. Weingarten, and U. Keller, Opt. Lett. 20, 1889 (1995). 7. N. Forget, A. Cotel, E. Brambrink, P. Audebert, C. Le Blanc, A. Jullien, O. Albert, and G. Chériaux, Opt. Lett. 30, 2921 (2005). 8. M. P. Kalashnikov, E. Risse, H. Schönnagel, and W. Sander, Opt. Lett. 30, 923 (2005). 9. A. Jullien, S. Kourtev, O. Albert, G. Chériaux, J. Etchepare, N. Minkovski, and S. M. Saltiel, Appl. Phys. B 84, 409 (2006). 10. F. Tavella, K. Schmid, N. Ishii, A. Marcinkevičius, L. Veisz, and F. Krausz, Appl. Phys. B 81, 753 (2005). 11. I. N. Ross, P. Matousek, M. Towrie, A. J. Langley, and J. L. Collier, Opt. Commun. 144, 125 (1997). 12. N. Ishii, L. Turi, V. S. Yakovlev, T. Fuji, F. Krausz, A. Baltuška, R. Butkus, G. Veitas, V. Smilgevičius, R. Danielius, and A. Piskarskas, Opt. Lett. 30, 567 (2005). 13. L. J. Qian, X. Liu, and F. Wise, Opt. Lett. 24, 166 (1999).
