We report on the first experimental measurement of high-dynamic-range pulse contrast of compressed optical parametric chirped-pulse-amplification (OPCPA) pulses on the picosecond scale. The measured -80-dB OPCPA contrast at 1054 nm agrees well with theoretical predictions and exceeds the estimated and measured levels for comparable amplification in a Ti:sapphire regenerative amplifier by approximately 10 dB. A key to achieving better contrast with OPCPA is the simpler experimental setup that promotes more-efficient coupling of seed pulse energy into the amplification system.
© 2007 Optical Society of America
Temporal pulse contrast is of foremost importance in high-intensity plasma-physics experiments where poor temporal contrast caused by prepulses and background pedestals can change the physical properties of targets before the arrival of the main laser pulse . High-energy, ultrashort laser pulses are typically generated in a mode-locked oscillator followed by chirped-pulse amplification (CPA) . The temporal contrast level typically associated with this technique is driven mainly by the amplifier noise or amplified spontaneous emission (ASE), resulting in a background intensity about 50 to 70 dB below that of the main pulse. When the laser is used to reach on-target intensities greater than 1018 W/cm2, the intensity background is high enough, on a long-enough scale, to change the target’s physical properties before the short pulse reaches it.
Two general approaches have been demonstrated to improve compressed pulse contrast: first, increasing the CPA amplifier seed improves the signal-to-noise ratio [3–5] and, second, nonlinear effects [6–9] can “clean” the short pulse from its pedestal after the pulse is temporally compressed. Both approaches add complexity to the system by adding one or more contrast improvement stages.
Optical parametric chirped-pulse amplification (OPCPA)  was proposed as a preamplification technique for the front end of large, petawatt-class Nd:glass laser systems because of its manifold advantages . In OPCPA, optical parametric generation (OPG) plays a role analogous to ASE in laser amplification. Since OPCPA involves an instantaneous transfer of power from a pump pulse to the amplified signal pulse, OPG is generated only during the pump-pulse duration. Lower prepulse energy can be achieved for target experiments with OPCPA, given that OPCPA pump pulses are only a few nanoseconds or less. Theoretical predictions that OPCPA should achieve higher contrasts  than laser amplification have been controversial , although a measured OPCPA contrast of 78 dB for nanosecond time scales has been reported .
In this paper, we report on the experimental characterization of the picosecond-scale temporal contrast of an OPCPA system operating at 1054 nm . In Sec. 2, we theoretically estimate the temporal contrast for an OPCPA system based on lithium triborate (LBO) as well as a Ti:sapphire regenerative amplifier. In Sec. 3, we describe the experimental setup for measuring the contrast of a compressed pulse amplified by OPCPA and, in Sec. 4, present the first direct measurement of high-dynamic-range temporal contrast of a 20-mJ pulse amplified using OPCPA. In Sec. 5, we compare our OPCPA results to the reported temporal contrast of a Ti:sapphire regenerative amplifier  and also compare the contrast limitations of each system.
2. Theoretical contrast of parametric and regenerative amplifiers
The level of parametric fluorescence in an optical parametric amplifier can be evaluated by assuming one photon of noise per mode and per time unit at the input of the amplifier and summing all amplified contributions over the relevant spectral and angular windows . To calculate the contrast, only the OPG that filters through the OPCPA system should be considered. This corresponds to the OPG emitted “on axis” with a spectrum included in the spectral bandwidth of the system. In such conditions, the OPG intensity emitted through the system is given by
where Δω sys is the spectral bandwidth of the system, Δθ sys is the divergence half-angle, and nc, vc, and λc are the index, group velocity, and wavelength of the fluorescence complementary wave, respectively. Equation (1) is valid only for the unsaturated amplification case, which provides a lower limit when significant pump depletion occurs. In addition, it is assumed that OPG emission is uniform across the considered spatial and spectral windows, which can be verified in Fig. 1, which shows the OPG power as a function of the wavelength and interaction angle for an LBO crystal cut for slightly noncollinear interaction. The yellow square represents the projection of the system’s angular and spectral windows and is included in an area of uniform OPG power.
The solid angle of the system accepts as a lower limit the divergence of the amplified beam (Δθ sys ~ λ/ω 0), but this limit is not easily reached for practical reasons. The spectrum of the OPG filtered by the system (laser amplifier, compressor) is more or less naturally matched to the amplified pulse spectrum [Δλs ~ λ 2/(Δtsπc)], so, in the following, we assume Δλs = Δλ sys. For a nearly degenerate configuration, a spectral width that is small compared to the wavelength and a system bandwidth that matches the OPCPA spectral width, Eq. (1) can be approximated as
To compare the OPG level to the signal level in terms of measurable input quantities, it is convenient to express the fluxes of OPG and signal photons. Assuming a pump pulse with a top-hat profile both in space and time, with area S, the equivalent flux of OPG photons emitted on axis is given by
When spectral gain variations can be neglected, the flux of amplified and compressed OPCPA signal photons is given by
where Es is injected signal energy, Δts is the ultimate pulse duration, and ωs is signal pulsation. Assuming perfect spatial and temporal overlap of the amplified OPCPA signal and OPG fluxes, the pulse contrast is then given by
This formulation of the pulse contrast can be understood as the product of three terms: the inverse of the number of injected photons, a term that characterizes the spatial discrimination between signal and OPG, and the time-bandwidth product of the recompressed pulse. Since Fourier-transform-limited pulses and good optical beam quality have been demonstrated with this OPCPA system , the theoretical limit for OPCPA pulse contrast is given by the number of injected photons.
In the case of an OPCPA injected with 0.6-nJ pulses, an OPCPA beam cross section of 3.4 mm2, a 1054-nm central wavelength compressed to 400 fs, and a compressor bandwidth of 10 nm, the contrast of a beam of 1-mrad divergence equals 82 dB when Eq. (5) is used.
The ASE from a regenerative amplifier also originates from the same quantum noise (one photon per mode), and its power can be found in Eq. (3) of Ref 13. The contrast can then be calculated from this formula and expressed in a form similar to Eq. (5)
where Δω is the spectral bandwidth of the regenerative amplifier, ω bd is the effective bandwidth of the fluorescence spectrum (~200 nm), hωs is the signal photon energy, σ is the stimulated emission cross section, τ is the lifetime of the metastable level, S is the area of the beam in the laser gain medium, and Ω is the small solid angle in which fluorescence is collected. Es and Δts stand for the energy and time duration of the injected pulse. For a TEM00 mode, the product of the beam area by the divergence of the beam is SΩ = λ 2/n 2:
Equations (5) and (7) both show a dependency on the number of injected photons; however, with OPCPA, this number is close to the available seed photons, while injection into a regenerative amplifier depends on many parameters and cannot easily be optimized. For the numerical application, the bandwidth of the Ti:sapphire regenerative amplifier operating at 1057 nm considered in Sec. 5 is 7 nm and limited by the transmission of the intracavity polarizer. The effective excited state lifetime τ is 2.8 μs. Despite careful mode matching, the fraction of energy effectively coupled to the Gaussian TEM00 mode of the cavity within the bandwidth of the regenerative cavity is estimated to be 20%, at best; therefore, the seed energy Es does not exceed 80 pJ. With these parameters, the expected contrast for the regenerative amplifier is 74 dB.
3. Experimental setup
Figure 2 shows a schematic of the experimental setup, but a more detailed description of the OPCPA system can be found in Ref. 11. A Nd:glass mode-locked oscillator (GLX-200, Time-Bandwidth Product) generates 220-fs, 1.7-nJ pulses at a repetition rate of 76 MHz and a wavelength of 1053 nm. The short pulses are temporally stretched in an Öffner triplet stretcher  by a stretch ratio of 300 ps/nm. The stretcher offers a spectral bandwidth slightly greater than 8-nm FWHM with an energy transmission efficiency of approximately 40%. The stretched 2.4-ns pulse is then sent to the OPA stage for amplification, which is the two-crystal preamplifier described in Ref. 15.
The seed beam is up-collimated to 2 mm (FWHM) to create a good overlap between the pump beam and the seed beam in the OPCPA crystal. Because of the instantaneous nature of parametric amplification, we use a 2.5-ns-long, top hat pump laser , running at 5 Hz, to provide a uniform gain across the pumped area in space and time, allowing for high conversion efficiency, without spatiotemporal effects. The amplifier uses two 29.75-mm-long lithium triborate (LBO) crystals in a walk-off compensated configuration and cut for nearly collinear, degenerate interaction. The pump beam is 2.2 mm in diameter, and its energy can be tuned to reach saturation of the parametric process, typically with 100-mJ pump pulses, at a pump laser intensity of 1 GW/cm2. With this amplifier, the measured gain exceeds 103 per crystal, and the output energy, after the two crystals, is greater than 20 mJ, leading to an overall gain of 3 × 107. The pump and seed beams are slightly noncollinear to allow for separation of the idler and signal beams. At nominal pumping power, the energy balance of the system defined at the sum of the idler, signal, and transmitted pump energy equals 90% of the input pump energy, indicating that 10% of the energy is either absorbed, reflected, or dissipated in form of OPG in the nonlinear crystals.
After amplification to 20 mJ, about 10% of the energy is sent through a diagnostic compressor. The 2-mJ pulse compresses to within 7% of the theoretical limit in the folded two-grating compressor, and the high-dynamic-range temporal profile shown in Fig. 2 is measured using a third-harmonic cross-correlator (Sequoia, Amplitude Technologies). The 10-mm-wide beam is not imaged throughout the compressor and undergoes an ~8-m free-space propagation before entering the cross correlator. The beam is not focused into the correlator crystal, and we can suppose that the entire beam participates in the measurement (no aperturing occurs).
The cross-correlation trace presented in Fig. 3 shows typical experimental data. The noise level of the cross correlator is assessed by blocking each of the arms of the apparatus during a scan. When the 2ω channel is blocked, as shown in Fig. 2, the noise level of -87 dB is higher than when the 1ω arm is blocked or when the input beam is totally blocked. In that case, the noise level drops to -110 dB, which is the detection noise floor. The noise floor of the cross correlator is apparently limited by self-third-harmonic generation of the 1ω arm due to scattering of the beam to a dynamic range of -87 dB.
4. High-dynamic-range OPCPA contrast measurements and analysis
Four distinct regions of contrast can be identified in the compressed OPCPA pulse temporal profile shown in Fig. 3: a transform-limited pulse, a pedestal divided into two regions with different slopes, and a constant pedestal for delays greater than approximately 100 ps before the main pulse. The Fourier-transform-limited, 400-fs pulse is characterized by a contrast of -27 dB, and a slight asymmetry of the pulse is observed. A slower decrease of the pulse intensity is observed in the region between 2 ps and 25 ps before the pulse with a slope of 10 dB per 15 ps, while the pulse intensity decreases even slower with a slope of 10 dB per 25 ps in the range of -25 ps to -100 ps. At this point, the temporal contrast reaches a steady value around an average of -79 dB up to the maximum range of the measurement window at -170 ps.
The mode-locked seed oscillator contributes to part of the slow-rise-time pedestal. The seed pulses were characterized using a scanning, noncollinear, second-order autocorrelator with a maximum dynamic range of ~60 dB, which is mostly limited by scattering effects in the frequency-doubling crystal. The result of the autocorrelation trace is superimposed on the OPCPA cross-correlation trace in Fig. 4(b). These two plots were not recorded simultaneously, but they both exhibit a pedestal with a similar slope between +20 ps and -20 ps, although the pedestal levels are different. We later found that the oscillator pulse pedestal is present when the oscillator spectrum is tuned toward 1053 nm to match the central wavelength of LHG8 Nd:glass, as required for injection of a petawatt-class laser. When the oscillator central wavelength is allowed to shift to 1055 nm and above, we experimentally verified that this pedestal disappears. Moreover, when monitored over time, the level of the pedestal varied during the day. This shows that part of the pedestal seen in the OPCPA output pulse is likely to have been present before amplification and should not be attributed to an effect arising during the parametric amplification. This slow rise time of the temporal intensity of pulses produced by a mode-locked oscillator using a SESAM has been reported before  and corroborates our observations.
The -100-ps to -25-ps region of the slow-rise-time pedestal, before the pulse, is consistent with OPCPA contrast degradation resulting from temporal noise on the OPCPA pump pulse . This high-frequency temporal noise is transferred to the OPCPA spectrum of the chirped pulse, which causes poor recompression on a picosecond time scale. Further investigation is required to fully evaluate and confirm this contrast degradation mechanism.
The level of the background signal was monitored daily over one week and, again, one month later and showed a constant value. No specific spatial filtering was applied to the OPCPA beam, nor was it image relayed to the correlator. However, when the beam was apertured in the compressor, we saw a decrease of the measured noise floor to values between -86 dB and -89 dB. As this is very close to the detection limit of the apparatus, the data are not quantitatively reliable. However, this would tend to verify that by reducing the second term in Eq. (5), which is the allowed propagation solid angle, one can improve the contrast.
5. Comparison with regenerative amplification
A direct comparison of contrast achieved with equivalent OPCPA and regenerative amplification performance highlights several of the advantages of OPCPA. The LULI 100-TW laser front end uses a regenerative amplifier that performs similarly to our OPCPA setup. In this setup, a commercial Ti:sapphire mode-locked oscillator (Tsunami, Spectra-Physics) delivers 100-fs (FWHM) pulses at 1057-nm pulses with a bandwidth of 12 nm at a repetition rate of 82 MHz. Pulses are stretched in an Öffner triplet stretcher to 1 ns. The energy of the stretched pulses is 0.5 nJ, which is practically identical to the OPCPA seed injection level described in Sec. 4. Selected pulses are injected into a Ti:sapphire regenerative amplifier pumped by a frequency-doubled Nd:YAG laser (SureLite, Continuum) and amplified to 2 mJ with a 10-Hz repetition rate. Gain narrowing reduces the bandwidth of the amplified pulses to 5 nm. A pair of gratings is used to recompress the pulse to 350 fs and characterized using the same high-dynamic-range cross-correlator, as shown in Fig. 4(a). The temporal contrast behavior for regenerative amplification differs significantly from that measured for the OPCPA system. A pulse pedestal extending ~20 ps before and after the peak of the compressed regenerative amplifier pulses is attributed to residual spectral phase. This pedestal drops to a steady contrast level of -70 dB, which agrees well with the theoretical estimate for ASE discussed in Sec. 2. Although the cross-correlator measurement is limited to ~100 ps before the pulse, this ASE pedestal is expected to extend over several nanoseconds corresponding to the pump laser pulse length.
The measured background ASE level for the regenerative amplifier is 10 dB higher than the OPG-limited background measured for the OPCPA system, which agrees well with the theoretical predictions and shows that pulse contrast is primarily differentiated by the efficiency of coupling seed energy into the amplification system. Given the simpler experimental setup, achieving high contrast is more straightforward with OPCPA than regenerative amplification because cavity mode-matching is more stringent than simple spatial and temporal pulse-overlapping.
We report the first high-dynamic-range temporal measurements of an optical pulse amplified by OPCPA on a picosecond time scale and compare them with results obtained for with regenerative amplification. Our results show a -79-dB noise floor for OPCPA that is limited by OPG and is in good agreement with theoretical estimates, but an approximately 100-ps temporal pedestal was observed that is consistent with measurements reported in Ref. 21. On a nanosecond scale, OPCPA offers approximately 10-dB-better contrast than classical regenerative amplification, primarily due to better coupling of seed pulses into the amplification system.
Some solutions should be tested to improve the contrast of pulses amplified by OPCPA. First, the use of an OPCPA preamplifier stage, strongly spatially filtered, should improve the contrast as close to the theoretical limit as possible, and, second, the use of high-energy seed lasers, as was recently demonstrated, would improve the signal-to-noise ratio by 20 dB. With these two steps, it should be possible to improve the contrast of pulses amplified by OPCPA to -110 dB or -120 dB.
This work was supported by the U.S. Department of Energy Office of Inertial Confinement Fusion under Cooperative Agreement No. DE-FC52-92SF19460, the University of Rochester, and the New York State Energy Research and Development Authority. The support of DOE does not constitute an endorsement by DOE of the views expressed in this article.
References and links
1. D. Umstadter, “Review of physics and applications
of relativistic plasmas driven by ultra-intense
lasers,” Phys. Plasmas 8, 1774–1785
2. D. Strickland and G. Mourou, “Compression of amplified chirped
optical pulses,” Opt. Commun. 55, 447–449
4. A. Fernandez, T. Fuji, A. Poppe, A. Fürbach, F. Krausz, and A. Apolonski, “Chirped-pulse oscillators: A route
to high-power femtosecond pulses without external
amplification,” Opt. Lett. 29, 1366–1368
5. 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
6. A. Jullien, F. Augé-Rochereau, G. Chériaux, J.-P. Chambaret, P. d’Oliveria, T. Auguste, and F. Falcoz, “High-efficiency, simple setup for
pulse cleaning at the millijoule level by nonlinear induced
birefringence,” Opt. Lett. 29, 2184–2186
7. 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, “Temporal contrast in Ti:Sapphire
lasers: Characterization and control,”
IEEE J. Sel. Top. Quantum Electron. 4, 449–458
8. A. Jullien, O. Albert, F. Burgy, G. Hamoniaux, J.-P. Rousseau, J.-P. Chambarét, F. Augé-Rochereau, G. Cheriaux, J. Etchepare, N. Minkovski, and S. M. Saltiel, “10-10 temporal contrast
for femtosecond ultraintense lasers by cross-polarized wave
generation,” Opt. Lett. 30, 920–922
10. A. Dubietis, G. Jonusauskas, and A. Piskarskas, “Powerful femtosecond pulse
generation by chirped and stretched pulse parametric amplification in BBO
crystal,” Opt. Commun. 88, 437–440
11. L. J. Waxer, V. Bagnoud, I. A. Begishev, M. J. Guardalben, J. Puth, and J. D. Zuegel, “High-conversion-efficiency optical
parametric chirped-pulse amplification system using spatiotemporally shaped
pump pulses,” Opt. Lett. 28, 1245–1247
12. I. N. Ross, P. Matousek, M. Towrie, A. J. Langley, and J. L. Collier, “The prospects for ultrashort pulse
duration and ultrahigh intensity using optical parametric chirped pulse
amplifiers,” Opt. Commun. 144, 125–133
13. V. V. Ivanov, A. Maksimchuk, and G. Mourou, “Amplified spontaneous emission in a
Ti:Sapphire regenerative amplifier,”
Appl. Opt. 42, 7231–7234
14. H. Yoshida, E. Ishii, R. Kodama, H. Fujita, Y. Kitagawa, Y. Izawa, and T. Yamanaka, “High-power and high-contrast optical
parametric chirped pulse amplification by
crystal,” Opt. Lett. 28, 257–259
15. V. Bagnoud, I. A. Begishev, M. J. Guardalben, J. Puth, and J. D. Zuegel, “5-hz, >250-mJ optical
parametric chirped-pulse amplifier at 1053 nm,”
Opt. Lett. 30, 1843–1845
16. J. Zou, D. Descamps, P. Audebert, S. D. Baton, J. L. Paillard, D. Pesme, A. Michard, A. M. Suautivet, H. Timsit, and A. Migus, “LULI 100-TW Ti:Sapphire/Nd:Glass laser: A first step toward a high-performance petawatt facility,” in Third International Conference on Solid State Lasers for Application to Inertial Confinement Fusion, H. Lowdermilk, ed. (SPIE, Bellingham, WA, 1999), Vol. 3492, pp. 94–97.
17. A. Yariv, Quantum electronics, 3rd ed. (Wiley, New York, 1989).
18. G. Chériaux, P. Rousseau, F. Salin, J. P. Chambaret, B. Walker, and L. F. Dimauro, “Aberration-free stretcher design for
ultrashort-pulse amplification,” Opt.
Lett. 21, 414–416
19. V. Bagnoud, M. J. Guardalben, J. Puth, J. D. Zuegel, T. Mooney, and P. Dumas, “High-energy, high-average-power
laser with Nd:YLF rods corrected by magnetorheological
finishing,” Appl. Opt. 44, 282–288
20. I. D. Jung, F. X. Kaertner, J. Henkmann, G. Zhang, and U. Keller, “High-dynamic-range characterization
of ultrashort pulses,” Appl. Phys. B B65, 307–310
21. N. Forget, A. Cotel, E. Brambrink, P. Audebert, C. Le Blanc, A. Jullien, O. Albert, and G. Chériaux, “Pump-noise transfer in optical
parametric chirped-pulse amplification,”
Opt. Lett. 30, 2921–2923