Direct comparison of kilohertz-and megahertz-repetition-rate femtosecond damage threshold

Abstract : We performed femtosecond laser-induced damage threshold (fs LIDT) measurements with substantially different repetition rate Ti:sapphire laser systems: a 1 kHz regenerative amplifier and a 4.3 MHz long-cavity oscillator. All other pulse parameters are kept the same. Comparative measurements of a dielectric high reflector, a chirped mirror, and metallic mirrors show at least a factor of 2.7 lower fs LIDT at megahertz repetition rates. We attribute this to thermally assisted damage mechanisms supported by complex heat transfer simulations. In recent years, the research of femtosecond laser-induced damage threshold (fs LIDT) has grown fast as different types of femtosecond lasers have become widely available. One can distinguish single-shot and multishot LIDT measurements. Multishot studies are typically limited to repetition rates in the hertz–kilohertz (Hz–kHz) domain with different samples [1–16], providing a huge amount of useful data for femtosecond laser developers up to kilohertz (kHz) repetition rates. Data on high reflectors (HRs) versus chirped mirrors (CMs) [8], round-robin experiments with different facilities [9], and single measurements on different types of HRs and CMs [10] are all available. Systematic repetition rate dependence comparisons are also abundant in the Hz–kHz range, showing results either without any significant changes or trends in fs LIDT [17,18] or with a slightly decreasing LIDT trend toward a higher repetition rate [19]. However, due to obvious reasons, in the past development of femto-second laser technology, kilohertz–megahertz (kHz– MHz) damage comparisons have not been targeted. As a parallel development, MHz-repetition-rate femto-second lasers with relatively high pulse energy have emerged recently, e.g., passively mode-locked Yb thin-disk lasers [20–22] and femtosecond fiber chirped pulse amplification systems [23]. Therefore, it is important to investigate the fs LIDT in this regime and directly compare it to standard kHz results. A measurement with a 100 MHz repetition rate oscillator on a HR mirror with a very tightly focused beam [24] and a round-robin measurement comparing this result with kHz-repetition-rate measurements have been reported, but the drawback is that several parameters of the sample illumination differ significantly [25]. Angelov et al. presented kHz–MHz comparison with an unknown number of pulses using a picosecond source, observing up to a factor of 2 difference depending on the material bandgap [26,27]. Therefore, it is of utmost importance to perform a direct comparison fulfilling the following requirements: (i) it studies the femtosecond damage on relevant optics (ii) between kHz-and MHz-repetition-rate illumination and (iii) has every other pulse parameter well controlled and unchanged. To the best of our knowledge, our current Letter presents such results for the first time. To this end, we performed such measurements with kHz-and MHz-repetition-rate lasers under exactly the same conditions and performed simulations to understand the underlying mechanisms of the observed difference. We took special care to provide exactly the same conditions for damage, which includes fixing the sample, wavelength, focal spot size, pulse length, and number of interacting pulses and changing strictly only the repetition rate. For these tests, we used two laser systems. First, we utilized a Ti:sapphire regenerative amplifier (RA) operating at 1 kHz repetition rate with 795 nm central wavelength , then we used a home-built long-cavity Ti:sapphire oscillator (LCO) at 4.3 MHz repetition rate with 805 nm central wavelength. Otherwise, the pulse parameters used for the damage tests are the same for both cases. The pulse length was 120 10 fs, which was the transform limited output of the LCO. In the case of the RA, this pulse length was achieved by chirping the 40 fs amplifier output to 120 fs with the built-in compressor grating pair. The 1∕e 2 diameter of the beam focused to the sample was 9.8 0.5 μm, achieved by an 18 mm focal length aspheric lens in both cases [Fig. 1(a)]. When we consider possible displacement of the sample with respect to the focus, we arrive at a spot size deviation of 0.2 μm, which leads to a smaller error than the precision of the diameter measurement. Different power levels were set by a wheel density filter. The number of pulses interacting with the sample was adjusted with a shutter with a minimum opening time of around 9 ms, which led to approximately 35,000 pulses of the LCO on the sample. Exactly the same number of pulses was applied on the sample with the shutter in the case of the RA, too. The damage event was detected visually with a stereomicroscope with 3× magnification and a CCD camera as shown in Figs. 1(b)–1(e). As the damage spot was very small and only visible on a few pixels, the method used was not able to determine in situ what kind of permanent alteration was caused. An independent measurement with a white-light inter-ferometric microscope confirmed that these spots were ablations in all cases. We applied normal incidence and the sample was placed on an xyz translation stage.
Document type :
Journal articles
Complete list of metadatas

Cited literature [24 references]  Display  Hide  Download
Contributor : Laurent Gallais <>
Submitted on : Wednesday, November 18, 2015 - 7:49:52 AM
Last modification on : Monday, March 4, 2019 - 2:04:24 PM




B.J. Nagy, Laurent Gallais, L Vámos, D Oszetzky, P Rácz, et al.. Direct comparison of kilohertz-and megahertz-repetition-rate femtosecond damage threshold. Optics Letters, Optical Society of America, 2015, 40 (11), pp.2525. ⟨10.1364/OL.40.002525⟩. ⟨hal-01228360⟩



Record views