Monolithic, mode-locked Titanium-Sapphire lasers with repetition rates in the range of 30 GHz to 300 GHz
Final Report Abstract
In this DFG project, monolithic titanium-sapphire (Ti:Sa) lasers were built. The aim was to generate mode-locked pulses with repetition rates from 30 GHz to 300 GHz. Planar Ti:Sa disks with thicknesses of 0.3 mm to 3 mm are used as laser crystals, which are dielectrically coated on both surfaces. These layers form the laser resonator. The coatings are chirped to compensate for the second order dispersion of the laser crystal. The Ti:Sa crystal is pumped with a frequency tunable single-frequency pump laser at 532 nm, inducing a thermal lens in the crystal which stabilizes the plan-plan resonator. Mode locking should be self-starting by using soft-aperture Kerr-lens mode locking as the gain becomes more saturated due to self-focusing. A total of four monolithic Ti:Sa lasers with 3 mm resonator length and two with 1 mm resonator length were set up. Unfortunately, mode locking could not be achieved. According to our current knowledge, there are three possible reasons for this: Frequency pulling due to the standing wave of the pump beam within the resonator generated by the double pass of the pump light through the crystal, spatial hole burning, and insufficient dispersion compensation. The standing wave of the pump light creates a periodic structure of the gain, causing frequency pulling of the longitudinal modes of the laser. This is problematic since the fixed phase relationship of the longitudinal modes during mode locking requires equal spectral mode spacings. In the future, a longer crystal (e.g., 5 mm) could be used, since for sufficient absorption a single pass of the pump beam would be enough. Our laser also exhibits strong spatial hole burning of the longitudinal modes due to the monolithic structure, which increases the bandwidth of the laser up to 10 nm in cw operation. Insufficient dispersion compensation could then cause a lower bandwidth while mode locking, making the gain less saturated. Limiting the bandwidth of the laser to for example 5 nm using the coatings would make the accuracy of the dispersion compensation less critical. The cw operation of the laser is very stable. The power variation of the laser over one hour was less than 0.04% RMS. The beam quality is good with an M² of less than 1.05 and the alignment of the laser is not critical. Such a stable and broadband cw laser could be used for optical coherence tomography. Autocorrelation measurements showed pulse-like signals caused by white light interference of the spectrally broad cw laser radiation. The measurements may falsely suggest a mode-locked pulse. Since this problem is not widely known, we are currently preparing a publication on this subject.
Publications
-
"Towards a monolithic, multi-gigahertz mode-locked Ti:Sa laser," Europhoton, Hannover, Poster (2022).
T. Fiehler & U. Wittrock
