796
OPTICS LETTERS / Vol. 31, No. 6 / March 15, 2006
Long-range filamentary propagation of subpicosecond ultraviolet laser pulses in fused silica S. Tzortzakis Laboratoire d’Utilisation des Lasers Intenses, Ecole Polytechnique, Centre National de la Recherche Scientifique; Unité Mixte de Recherche 7605, 91128 Palaiseau, France
D. G. Papazoglou Foundation for Research and Technology—Hellas, Institute of Electronic Structure and Laser, P.O. Box 1527, Heraklion 711 10, Greece, and Department of Materials Science and Technology, University of Crete, P.O. Box 2208, Heraklion 710 03, Greece
I. Zergioti Department of Physics, National Technical University of Athens, Iroon Polytexneiou 9, 15780 Zografou, Athens, Greece Received October 3, 2005; revised November 25, 2005; accepted November 29, 2005; posted December 5, 2005 (Doc. ID 65135)
We report on what is to our knowledge the first observation of subpicosecond ultraviolet laser pulse filamentation in transparent solid materials. Robust filaments were created in fused silica and observed over distances that exceed 3 cm in length. Intensities as high as 1013 W / cm2 were found to be transported in the filamented beam without material damage in the bulk of the fused-silica sample. © 2006 Optical Society of America OCIS codes: 190.5530, 190.5940.
Modern laser systems that deliver ultrashort pulses find numerous applications today in basic and applied sciences. Basic studies include the understanding of physical processes on time scales that permit the observation of electron motion, with applications in femtochemistry1 and biology. Other applications include precise micromachining and the fabrication of nanocomponents.2 Also, important applications can be found from the generation of secondary sources driven by fs laser pulses. Such sources span the production of higher harmonics and x rays3 to mid-infrared sources and microwaves,4 most of them with the same short pulse characteristics as the initial driving laser pulse. Finally, a whole new world of physics has been opened by the generation of attosecond pulses driven by few-cycle fs laser systems.5 One of the most spectacular effects involving fs laser pulses is their filamentation, which appears during their propagation in nonlinear media. The dynamic interplay between Kerr self-focusing and higher-order nonlinearities6–8 has been shown to lead to the self-trapping of short and intense laser pulses. This phenomenon is known in the literature as filamentation of the laser pulse. The term “filamentation” has been adopted because it describes well the spatial confinement of the laser beam over distances that extend well beyond the characteristic Rayleigh length. Ultrashort laser pulse filamentation was observed in the atmosphere in 1995,9 and since then it has been studied under different conditions. Filamentation has been observed in gases by use of different laser wavelengths from the infrared to the ultraviolet. It also was recently observed in transparent solids by use of femtosecond infrared pulses.10 Filaments have common attributes: spatial confine0146-9592/06/060796-3/$15.00
ment, temporal breakup and pulse shortening,11 spectral broadening, and ionized strings are found in all filaments, whatever the laser source or the choice of medium. Numerous are also the applications of the filaments, from high energy and intensity deposition at long distances to atmospheric monitoring and laser guiding of electric discharges. In this Letter we report on what is to our knowledge the first experimental observation of ultraviolet filamentation in fused silica. Using high ultraviolet photons (5 eV), two-photon processes are enough for electrons to leap over the material bandgap (7.8 eV). One would thus imagine that important material damage would prevent pulse filamentation. Such is shown not to be the case; not only is filamentation observed but it extends over much longer lengths than in all previous observations of femtosecond filamentation in solids reported in the literature. For the observation of ultraviolet filamentation we used the setup shown in Fig. 1. A femtosecond excimer (KrF) oscillator–amplifier system in conjunction with a dye laser12 is used. It produces linearly polarized pulses at 248 nm with duration of 450 fs and as much as 20 mJ of energy at a repetition rate of a few hertz (typically used at 5 Hz). The laser system consists of a double chamber Lambda Physik excimer laser and an ultrashort dye laser system. The XeCl excimer laser oscillator beam pumps a series of dye lasers to produce a subpicosecond green (496 nm) laser pulse. These pulses are frequency doubled in a nonlinear -barium borate crystal and triply amplified in the KrF cavity of the excimer laser. The final laser beam had a square profile of 50 mm⫻ 50 mm, a full divergence angle of 0.15 mrad, and a top-hat energy distribution. The central part of © 2006 Optical Society of America
March 15, 2006 / Vol. 31, No. 6 / OPTICS LETTERS
Fig. 1. Experimental setup for detection of the filament inside a fused-silica prism by a CCD camera equipped with a zoom lens and a red filter. Inset, spectral measurements of the laser pulse after propagation in the sample with a fibered spectrometer coupled to an intensified CCD camera.
the laser beam was selected by use of a 25 mm diameter iris to permit a uniform beam profile to be used in the experiment. The energy of the beam was adjusted by a variable attenuator. A focusing lens 共f = 300 mm兲 then gently converged the beam in a fused-silica sample, which had a prismatic form. The maximum propagation length was limited to 4 cm, the dimension of the longest edge of the prism. Finally, the fused-silica sample was positioned in such a way that the geometrical focus of the lens was at the center of it, as shown in Fig. 1. We monitored the propagation of the beam in the sample by recording the red fluorescence of the fused silica, centered at 650 nm. This fluorescence is known to result from a two-photon-induced color-center formation in fused silica.13 We recorded the fluorescence by placing a 10 bit linear CCD camera, equipped with an adapted zoom lens, perpendicular to the sample (Fig. 1). The chosen zoom factor allowed the whole propagation path in the sample, i.e., the 4 cm edge, to be recorded in a single frame. The spatial resolution of the images was better than 40 m. A red bandpass filter was placed in front of the CCD lens system to limit the surrounding noise and thus produce betterquality images of the filament. A critical parameter that one has to take into account is the energy losses that are due to two-photon absorption in fused silica.14 We measured the losses in our fused-silica sample and found that for input energies ranging from 1.0 J to 4.0 mJ the exit energy was 1.0 J to 0.6 mJ. As we have shown with numerical simulations of the propagation that take two-photon absorption into account, the greatest energy losses take place at the early stages of light propagation in fused silica, in the first few millimeters. Filamentary propagation images as a function of the input laser energy were recorded and are shown in Fig. 2. One can clearly observe the increased length of the spatially confined filamentary propagation as the input laser energy increases. The filament diameter, within the limits of our spatial resolution, is constant with input energy. Filamentation was observed with a single shot; yet for the images shown in Fig. 2, we accumulated 20 pulses in the CCD camera to obtain a comfortable signal-to-noise ratio, as the fluorescence signal was faint.
797
Except from the energy-dependent length of the filament one can also observe a considerable shift of the linear focus, owing to self-focusing, toward the laser source. A closer look at the image taken at the highest energy shows that filamentation starts practically after a few-millimeter propagation in the sample and has a length of more than 3 cm (almost 200 times the Rayleigh length). This is, to our knowledge, the longest femtosecond laser filament ever reported in solids. Actually, at the early stages of the propagation more than one filament, which later fuse, might exist, a behavior well known from previous studies of filamentation. The spatial resolution of our imaging system and the blurring that is due to the accumulation of many pulses in a single frame did not permit us to distinguish clearly between the multiple filaments. As we mentioned above, one would be cautious about ultraviolet laser filamentation in solids owing to the low-intensity damage thresholds. In addition, considering the initial pulse duration, the measured output energy, and the smallest beam diameter, we estimate that the intensity in the filamented beam can exceed 1013 W / cm2. (Although the filament actually contains less energy, at the same time its pulse duration is shorter, leading to similar intensities.) Nevertheless, filamentation is observed and at the same time the sample seems to resist the high intensities after the accumulation of an important number of shots on the same area. In fact, the only limiting parameter in our experiments was surface deterioration, which appeared after a few tens of shots (at the highest input energies), but even then no damage was detectable in the bulk of the sample. One plausible explanation for the observed material resistance would be the significant laser pulse shortening in the filamented pulse, which is one of the general filament attributes10 that would result in a limited number of electron avalanches. Actually, electron avalanche has been shown to be the key mechanism for bulk material damage with femtosecond laser pulses.15 The electron density level has been shown to be directly related to the damage produced in the bulk of the transparent solid, and its limits were found to be in the range 2–4 ⫻ 1020 cm−3. It has also been demonstrated numerically that these values of electron density can be attained only through electron avalanches.15 The exact physical
Fig. 2. (a) Filament images as a function of input laser energy. Laser intensity in gray scale: White represents the most intense. (b) Filament length as a function of input energy.
798
OPTICS LETTERS / Vol. 31, No. 6 / March 15, 2006
mechanisms involved in this effect for our experiments are under investigation, as their determination is of great importance for applications in the transfer of high laser energies and intensities through transparent media. Recently it was shown16 that structural modifications can be induced in fused silica when the same laser pulses as those used here are tightly focused. In this case avalanche ionization should play an important role, as was the case in Ref. 15. Filamentation is always accompanied by a moreor-less important spectral broadening, depending on the laser and material properties. To verify this attribute for the filaments of the present study, we recorded their power spectrum. The total internal reflection of the filamented beam on the prism’s exit surface was directed through an optical fiber in a visible spectrograph equipped with a 16 bit cooled intensified CCD camera (Fig. 1, inset). The initial laser power spectrum is shown in Fig. 3, represented by the continuous curves. Also in the same figure is shown, by dashed curves, the spectrum of the filament at 4.13 mJ input energy. Clear pulse broadening is observed for the filamented pulse in accordance with previous reports on filaments in the literature. Also, the broadening is of the same order of magnitude as for the ultraviolet filaments in air.17 Such was also the case for infrared filaments in air and solids.10 At low input energies of ⬃1 J, at which no filamentation is observed, the spectral broadening is much smaller [dotted curve in Fig. 3(a)]. Finally, in Fig. 3(b) are plotted the power spectra for filaments at different input energies. The filament spectra show almost the same broadening and profiles despite the differences in the initial energies; this is the result of intensity clamping and similar temporal characteristics in all filaments. In conclusion, we have presented what is to our knowledge the first experimental demonstration of ultrashort ultraviolet laser pulse filamentation in fused silica. The observed filaments have the same attributes found in filamentation in general, confirming the universal nature of the phenomenon, which is independent of the initial laser wavelength or propagation medium used. One of the most surprising observations was the resistance of the material to ex-
Fig. 3. Power spectra of the initial laser pulse and the pulse after propagation through the fused-silica sample as a function of input laser energy: (a) comparison of linear and filamented propagation, (b) filamentation at several input energies.
tremely high energies and intensities, which could be the result of an important pulse shortening in the filamented pulse. The last-named attribute could open the way for applications when the transfer of high laser energies and intensities through solids is necessary. The authors gratefully acknowledge the assistance of A. Eglezis, G. Sgouros, G. Maravelias, and S. Christopoulos. Experiments were carried out at the Ultraviolet Laser Facility operating at the Institute of Electronic Structure and Laser of the Foundation for Research and Technology—Hellas and supported by the European Union through Laserlab-Europe (RII3-CT-2003-506350). This work was also partially financed by the projects ENTER 01-ER-68 and PYTHAGORAS-I 68/829 (cosponsored by European and Greek funds). S. Tzortzakis’s e-mail address is
[email protected]. References 1. M. Dantus, and A. Zewail, Chem. Rev. (Washington, D.C.) 104, 1717 (2004). 2. K. M. Davis, K. Miura, N. Sigumoto, and K. Hirao, Opt. Lett. 21, 1729 (1996). 3. R. Bartels, S. Backus, E. Zeek, L. Misoguti, G. Vdovin, I. P. Christov, M. M. Murnane, and H. C. Kapteyn, Nature 406, 164 (2000). 4. S. Tzortzakis, G. Méchain, G. Patalano, Y.-B. André, B. Prade, M. Franco, A. Mysyrowicz, J.-M. Munier, M. Gheudin, G. Beaudin, and P. Encrenaz, Opt. Lett. 27, 1944 (2002). 5. P. M. Paul, E. S. Toma, P. Breger, G. Mullot, F. Augé, Ph. Balcou, H. G. Muller, and P. Agostini, Science 292, 1689 (2001). 6. M. Mlejnek, E. M. Wright, and J. V. Moloney, Opt. Lett. 23, 382 (1998). 7. A. Couairon, Eur. Phys. J. D 27, 159 (2003). 8. A. Dubietis, E. Gaižauskas, G. Tamošauskas, and P. Di Trapani, Phys. Rev. Lett. 92, 253903 (2004). 9. A. Braun, G. Korn, X. Liu, D. Du, J. Squier, and G. Mourou, Opt. Lett. 20, 73 (1995). 10. S. Tzortzakis, L. Sudrie, M. Franco, B. Prade, A. Mysyrowicz, A. Couairon, and L. Bergé, Phys. Rev. Lett. 87, 213902 (2001). 11. C. P. Hauri, W. Kornelis, F. W. Helbing, A. Heinrich, A. Couairon, A. Mysyrowicz, J. Biegert, and U. Keller, Appl. Phys. B 79, 673 (2004). 12. S. Szatmari and F. P. Schafer, Appl. Phys. B 46, 305 (1989). 13. F. G. Omenetto, W. A. Schroeder, K. Boyer, J. W. Longworth, A. McPherson, and C. K. Rhodes, Appl. Opt. 36, 3421 (1997). 14. D. N. Nikogosyan, Properties of Optical and Laser Related Materials: a Handbook (Wiley, 1997). 15. L. Sudrie, A. Couairon, M. Franco, B. Lamouroux, B. Prade, S. Tzortzakis, and A. Mysyrowicz, Phys. Rev. Lett. 89, 186601 (2002). 16. D. G. Papazoglou, I. Zergioti, S. Tzortzakis, G. Sgouros, G. Maravelias, S. Christopoulos, and C. Fotakis, Appl. Phys. A 81, 241 (2005). 17. S. Tzortzakis, B. Lamouroux, A. Chiron, S. Moustaizis, D. Anglos, M. Franco, B. Prade, and A. Mysyrowicz, Opt. Commun. 197, 131 (2001).
