We demonstrate for the first time that up to 0.45 µJ pulses can be obtained from a cavity-dumped Ti: sapphire oscillator stably operating in the positive dispersion regime. The output pulse can be compressed to 60 fs and used to generate a white light continuum through self-filamentation in a thin sapphire plate.
©2006 Optical Society of America
Cavity dumping is a standard technique for generating moderate-energy pulses from femtosecond Ti: sapphire laser oscillators [1–4]. A cavity dumper can increase the effective output coupling from a Ti: sapphire oscillator from 5–10% up to ~80%, giving an order of magnitude increase in pulse energy from the laser, at the expense of reduced repetition-rate. Pulse energies of >200 nJ have been reported from electro-optically cavity dumped Ti:sapphire oscillators , but at very low repetitions rates (kHz) and with limited repeatability. More generally, multi-pulse and continuous wave (CW) breakthrough instabilities from excessive Kerr nonlinearity have limited the obtainable stable-and-repeatable energy to tens of nJ energy and ~ 100mW average power. Other materials such as Yb:glass  and Yb:KYW , have operated mode locked and cavity dumped by employing semiconductor saturable-absorber mirrors (SESAMs). Although µJ pulse energies can be generated by these lasers, the peak power is only a few MW since the pulse duration is limited to a few hundred fs. For applications in materials micromachining , nonlinear frequency conversion, and femtosecond white-light continuum generation , a MHz repetition-rate source with ~µJ pulse energy and sub-100 fs pulse duration would be desirable.
An alternative to cavity dumping is to extend the length of the laser cavity to reduce the repetition rate and increase the per-pulse energy [10–14]. The extreme example of this is to include a Herriott-type multi-pass telescope in the standard oscillator cavity to obtain repetition rates as low as 2 MHz , corresponding to a 150m cavity length. To avoid excessive Kerr nonlinearity due to the high intracavity peak power, these oscillators are operated either with large negative or small positive intracavity dispersion, and in some cases with SESAMs to support stable mode locking [12, 15, 16]. 43 fs, 150nJ pulses have been generated by working in the large negative dispersion regime (NDR). In the positive dispersion regime (PDR) , the pulse in the cavity is broadened by the combined action of positive group velocity dispersion (GVD) and self phase modulation (SPM), and the laser is mode locked by the dynamic gain saturation, spectral filtering, and self amplitude modulation (SAM) or Kerr-lens mode-locking (KLM) . In PDR, a heavily chirped ps pulse is generated that can be externally compressed to <100 fs [14, 15]. The long pulse duration inside the laser in the PDR offers much better energy scalability, with acceptable pulse quality. Recently 130nJ pulse at 11MHZ and 505nJ pulse at 2MHz has been obtained from these chirped-pulse multi-pass cavity (CPMC) oscillators, with peak power as high as 10MW and 1~2W average power using a 10W pump laser .
Here, we show for the first time that it is possible to combine cavity dumping with the positive dispersion regime of operation. Using a configuration similar to the standard (prism dispersion controlled) cavity-dumped geometry , sub-100fs pulses with peak power as high as 7.5MW at a variable repetition rate as high as 2 MHz can be generated stably and reliably. Compared with the CPMC oscillator, this positive dispersion cavity dumped (PDCD) oscillator is substantially more energy-efficient, producing up to 450nJ pulse using as low as 6.5 W pump power. Furthermore, it has an easily variable repetition-rate and occupies a footprint of only 0.3m2—no larger than a standard Ti: sapphire oscillator. This oscillator fills a gap between standard Ti:sapphire oscillators and more-complex oscillator-amplifier systems . Finally, we demonstrate stable filament-induced white light generation from 400nm to 1µm using this laser.
2. Laser setup and performance
The laser system setup and external double-pass prism pair compressor is shown in Fig. 1. A Bragg cell driver and 18W amplifier (CAMAC) drive a fused silica cavity dumper (Harris). The laser repetition-rate is detected by a photodiode, conditioned using a constant fraction discriminator (CFD, Phillips Scientific) and slightly amplified by a pulse generator (Hewlett-Packard 8082A), before going into the Bragg cell driver. All the mirrors are standard dielectric mirrors with low dispersion. The path length in the Ti: sapphire crystal is ~5 mm.
With a 5.3% output coupler, the laser can be mode locked stably in both the NDR and PDR. In the NDR, the pump power is limited to below 5W to avoid instabilities. For the transition from negative to positive dispersion, we simply add material dispersion by translating the first intracavity prism. Near the zero dispersion point, the laser shows self Q-switched mode-locking behavior, but with more positive dispersion it settles into a stable, hard-edge, spectrum characteristic of a standard Ti:sapphire oscillator operating in the PDR . Increasing the pump power to the full 10.4W once in the PDR, we then optimize the intracavity dispersion and the KLM parameters, i.e. the relative position of the Ti: sapphire crystal and the focusing mirrors. In translating the prism from the negative dispersion point, we estimate that stable operation with ~40 nm bandwidth is obtained with a net intracavity dispersion of less than 40fs2. This value is significantly smaller than previous reported using SF10 prisms for dispersion compensation in a regular oscillator , or using fused silica prisms in multi-pass cavity . The CW output from the output coupler is 2W when pumped by 10.4 W in the PDR, indicating an intracavity pulse energy of 500nJ. The crystal was cooled to -3 °C using antifreeze coolant, and we saw no evidence of thermal lens induced degradation of slope efficiency even at full pump power. Given the very high intracavity pulse energy, non-resonant re-absorption of the 780 nm circulating light might be an issue; however, this absorption appears to be minimal.
A further increase in the intracavity power can be obtained by replacing the output coupler with a standard broadband high reflector (Newport UF20). Stable mode locking operation is sensitive to the relationship between cavity adjustment and dispersion. By comparing the integration of the detected intracavity spectrum with the output coupler used, we estimate that up to 60W intra-cavity power and 750nJ per pulse energy can be obtained at pump power as low as 6.5 W. Stable mode locking, however, is not obtained for higher pulse energies. We believe the limitation is set by saturation of the Kerr lens induced differential gain in the Ti: sapphire crystal, which can occur when the spot size of the red light becomes smaller than the pump light spot size. Robust operation at near-optimum power can be obtained with a highreflector and 6.5 W pump power. We have used both a Spectra-Physics Millennia X and Coherent Verdi 6 pump laser with comparable results.
When the intracavity peak-power is increased by substituting the output coupler with a high reflector, the spectrum is further broadened, as shown in Fig. 2 (blue). In this regime, we could obtain up to 60% dumping efficiency without encountering instabilities such as period-doubling or self Q-switching. In the NDR, cavity-dumping only slightly shifts the laser spectrum. However, in the PDR, the spectrum width and shape can be modified dramatically at MHz dumping frequencies, as shown in Fig. 2 (red). This results from the fact that when the output coupler is replaced with a high reflector, the overall cavity loss (and therefore the intracavity pulse energy) is dominated by the cavity dumping. Furthermore, in the PDR, the Kerr lens is expected to play a more important role in determining the spectrum inside the cavity than it would for the standard NDR solitary mode of operation.
To obtain the highest pulse energy in a stable mode of operation, we used an RF spectrum analyzer (Agilent E4410B) to monitor the pulse train, and limited the Bragg driver power to below where additional amplitude noise is induced by cavity dumping. A typical microwave spectrum of the intracavity modulated pulse train is shown in Fig 3.
The dependence of the cavity dumped pulse energy on dumping frequency is shown in Fig. 4. The peak pulse energy is obtained at 0.8MHz; lower dumping rates give slightly lower energy than at 0.8 MHz, while higher dumping rates give proportionately lower pulse energy. The peak at 0.8 MHz results from relaxation oscillations that give an energy overshoot at about 1.25 µs after dumping of the previous pulse.
Using a fast photodiode (100 ps resolution), combined with a long scan-range (>100 ps) autocorrelator, and a high resolution spectrometer, we confirmed that no secondary pulses are generated even at the highest intracavity peak power. We also measured the intensity contrast ratio between the dumped output pulse and the neighboring pulse. The delay of the Bragg cell driver can be adjusted to give a<1:100 ratio for both the pre- and the post-pulse.
The uncompressed pulse emitted from the laser has duration of about 1.2ps. Using a simple compressor consisting of two equilateral SF11 prisms with 90cm separation, we recompressed the pulse with high (>90%) throughput. Double-passing the compressor gives approximately -10000 fs2 GVD and -50000fs3 third-order dispersion. A frequency-resolved optical gating (FROG) measurement of the recompressed pulse is shown in Fig. 5. The FWHM pulse duration is 60fs, is ~1.3x transform limited, with a peak power of ~7.5MW. The satellite ripples are generated by the sharp edge of the spectrum and the residual third-order dispersion of the prism compressor. The PDCD oscillator and CPMC oscillator have similar sharp edge spectra and satellite pulses in their temporal profiles [14, 15]. We observe slightly longer pulse duration (60fs) than in the case of the CPMC oscillator (45fs at 2MHz repetition rate), because our compressor can not compress the pulse to the transform limit of the spectrum (≈46 fs). Careful engineering of the intracavity dispersion can further broaden the spectrum. The use of a high-throughput pulse shaper or specially-designed TOD compensating chirp mirrors should result in nearly transform-limited pulse duration. As is generally found in lasers operating in the PDR [15, 17], the spectral width and chirp of the output pulse strongly depends on the intracavity dispersion, pulse energy and cavity alignment. Thus the extra cavity dispersion must be re-optimized when the operating parameters of the laser change.
To demonstrate the unique utility of this laser, we generated a white light continuum by focusing the compressed 0.8MHz repetition-rate pulse into thin sapphire and fused silica plates using a 6 cm FL aspheric lens. Filament-induced white light generation could be observed without the need for careful optimization. The use of sapphire avoids rapid damage and allows extended operation of the white light source. The white light spectrum measured using an optical spectrum analyzer (Ando AQ-6315E) is shown in Fig. 6. The inset of Fig. 6 is a digital camera image of the continuum. Although the center of the white light profile is stable, we observe a long time-scale, periodic (~10 seconds) instability in the red ring. We speculate that this results from a thermal cycling due to the relatively high repetition-rate and average power of this filament compared with past work. To our knowledge these are the first data presented on white-light continuum from a filament generated by the unamplified output directly from a sub-100 fs Ti: sapphire laser oscillator. One recent paper using multi-pass cavity oscillators operating in the PDR with much higher pump laser power, has also mentioned white light generation . Past work using a cavity-dumped Yb:KYW laser with an additional stage of spectrum broadening and re-compression before generating filamentation [6, 8], greatly complicated the setup compared with the one discussed herein.
In conclusion, we have demonstrated the first cavity-dumped Ti: sapphire oscillator working in the positive dispersion regime. This results in a dramatic enhancement in the average and peak output power from such a laser, while still maintaining pulse duration of well under 100 fs. We believe that this pulse energy is currently limited by saturation of the KLM, and optimum operation can be obtained using a relatively-modest 6 Watt pump laser. Further improvement in the pulse energy might be obtainable using a SESAM, either through an increase in intracavity pulse energy, through an increased dumping of the intracavity pulse before instabilities arise, or through more freedom to optimize the power without losing mode locking. It is likely that energies exceeding 1 µJ can be obtained, and possibly much more with aggressive cooling of the crystal . Such a compact and reliable high peak-power laser could be used for a variety of scientific and industrial applications such as micromachining, nonlinear frequency conversion, broad-band fs spectroscopy, and even strong field physics experiments. Carrier-envelope stabilization of such a laser oscillator, or using it as a seed source for CW-pumped active amplification [20, 21] to the tens of µJ level at MHz repetition rate may also be possible.
This work was funded by the Office of Naval Research MURI program. We thank Dirk Müller at Kapteyn-Murnane Laboratories Inc. for assistance at the initiation of this project.
References and links
4. A. Baltuska, Z. Wei, M. S. Pshenichnikov, D. A. Wiersma, and R. Szipöcs, “All-solid-state cavity-dumped sub-5-fs laser,” Appl. Phys. B 65, 175–188 (1997). [CrossRef]
6. A. Killi, A. Steinmann, J. Dorring, U. Morgner, M. J. Lederer, D. Kopf, and C. Fallnich, “High-peak-power pulses from a cavity-dumped Yb : KY(WO4)2 oscillator,” Opt. Lett. 30, 1891–1893 (2005). [CrossRef] [PubMed]
7. A. M. Kowalevicz, V. Sharma, E. P. Ippen, J. G. Fujimoto, and K. Minoshima, “Three-dimensional photonic devices fabricated in glass by use of a femtosecond laser oscillator,” Opt. Lett. 30, 1060–1062 (2005). [CrossRef] [PubMed]
8. A. Killi, A. Steinmann, G. Palmer, U. Morgner, H. Bartelt, and J. Kobelke, “Megahertz optical parametric amplifier pumped by a femtosecond oscillator,” Opt. Lett. 31, 125–127 (2006). [CrossRef] [PubMed]
9. A. Brodeur and S. L. Chin, “Ultrafast white-light continuum generation and self-focusing in transparent condensed media,” J. Opt. Soc. Am. B 16, 637–650 (1999). [CrossRef]
10. S. H. Cho, B. E. Bouma, E. P. Ippen, and J. G. Fujimoto, “Low-repetition-rate high-peak-power Kerr-lens mode-locked Ti: Al2O3 laser with a multiple-pass cavity,” Opt. Lett. 24, 417–419 (1999). [CrossRef]
11. A. R. Libertun, R. Shelton, H. C. Kapteyn, and M. M. Murnane, “A 36 nJ - 15.5 MHz extended-cavity Ti:Sapphire Oscillator,” in CLEO 1999 Technical Digest: Optical Society of America, 1999.
12. S. H. Cho, F. X. Kärtner, U. Morgner, E. P. Ippen, J. G. Fujimoto, J. E. Cunningham, and W. H. Knox, “Generation of 90-nJ pulses with a 4-MHz repetition-rate Kerr-lens mode-locked Ti : Al2O3 laser operating with net positive and negative intracavity dispersion,” Opt. Lett. 26, 560–562 (2001). [CrossRef]
13. A. M. Kowalevicz, A. T. Zare, F. X. Kärtner, J. G. Fujimoto, S. Dewald, U. Morgner, V. Scheuer, and G. Angelow, “Generation of 150-nJ pulses from a multiple-pass cavity Kerr-lens mode-locked Ti : Al2O3 oscillator,” Opt. Lett. 28, 1597–1599 (2003). [CrossRef] [PubMed]
14. A. Fernandez, T. Fuji, A. Poppe, A. Furbach, F. Krausz, and A. Apolonski, “Chirped-pulse oscillators: a route to high-power femtosecond pulses without external amplification,” Opt. Lett. 29, 1366–1368 (2004). [CrossRef] [PubMed]
15. S. Naumov, A. Fernandez, R. Graf, P. Dombi, F. A. Krausz, and A, “Approaching the microjoule frontier with femtosecond laser oscillators,” New J. Phys. 7, 216 (2005). [CrossRef]
16. V. L. Kalashnikov, E. Podivilov, A. Chernykh, S. Naumov, A. Fernandez, R. Graf, and A. Apolonski, “Approaching the microjoule frontier with femtosecond laser oscillators: theory and comparison with experiment,” New J. Phys. 7, 217 (2005). [CrossRef]
18. S. Backus, C. Durfee, M. M. Murnane, and H. C. Kapteyn, “High power ultrafast lasers,” Rev. Sci. Instrum. 69, 1207–1223 (1998). [CrossRef]
19. S. Dewald, T. Lang, C. D. Schröeter, R. Moshammer, J. Ullrich, M. Siegel, and U. Morgner, “Ionization of noble gases with pulses directly from a laser oscillator,” Opt. Lett. 31, 2072–2074 (2006). [CrossRef] [PubMed]
20. S. Backus, R. Bartels, S. Thompson, R. Dollinger, H. C. Kapteyn, and M. M. Murnane, “High-efficiency, single-stage 7-kHz high-average-power ultrafast laser system,” Opt. Lett. 26, 465–467 (2001). [CrossRef]
21. R. Huber, F. Adler, A. Leitenstorfer, M. Beutter, P. Baum, and E. Riedle, “12-fs pulses from a continuous-wave-pumped 200-nJ Ti : sapphire amplifier at a variable repetition rate as high as 4 MHz,” Opt. Lett. 28, 2118–2120 (2003). [CrossRef] [PubMed]