Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Group delay dispersion tuned femtosecond Kerr-lens mode-locked Ti:sapphire laser

Open Access Open Access

Abstract

We report on a new design for a femtosecond Ti:sapphire oscillator in which dispersion compensation is realized exclusively using mirrors, including special mirrors with third order dispersion. This makes the oscillator dynamically tunable over a spectral range of 45 nm using an intracavity wedge-pair; and it delivers 40 fs pulses at 80 MHz repetition rate. Due to the all-mirror design, the oscillator represents an attractive base for a tunable frequency comb for high precision spectroscopy applications.

© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

1. Introduction

Utilizing the broad gain bandwidth of Ti:sapphire crystals, pulsed oscillators tunable over a broad spectral range [16] were realized or ultrashort pulse durations down to 10 fs [7,8] and far below [912] were reached. Such properties made these lasers attractive sources for spectroscopy, sensing, non-linear microscopy, coherent Raman imaging and coherence tomography applications. Furthermore, studying ultrafast processes with high temporal resolution and building frequency combs for high precision spectroscopy became possible.

Pulsed oscillators require suitable, low loss optical elements with negative intracavity group delay dispersion (GDD) to compensate the positive GDD of the used laser crystal and additional intracavity optical elements. Broadband negative GDD multilayer mirrors are one of the possible solutions. Such oscillators support the mentioned ultrashort pulse durations; they are not sensitive to short- and long-term beam directional stability and consequently they are excellent candidates to realize frequency combs [13] with repetition rate and carrier envelope phase (CEP) stabilization. They only drawback so-far was that they were able to work only on a particular central wavelength, and they were not tunable. This study shows how to overcome this limitation.

Usually, to realize tunable laser oscillators, Brewster prism-pair based compressors are placed into the cavity, which produces the necessary negative GDD and at the same time, serves as a spectral filter to select and tune the central wavelength by an aperture slit. Spatial chirp or TOD might contribute to the tuning of the oscillator presented in [3], however it was not stated or discussed in the paper. With such oscillators, tuning ranges up to 100-400 nm were reached [1,3,5,6] at 50-150 MHz repetition rates with pulse durations below 100 fs. At a low repetition rate of 22 MHz, 170 nm tuning range and 150-300 fs pulse duration was demonstrated [4].

Other realizations utilized birefringent filters together with specially designed broadband chirped mirrors [2] and an oscillator with 70-fs-long pulses was realized with a tunability in the 800 nm to 900 nm spectral range. Synchronously pumped [14] or actively mode-locked [15] oscillators can be tuned by exploiting that the first order dispersion, or group delay and consequently the round-trip time, is wavelength dependent. Changing the cavity length, the wavelength was forced to adapt to get the round-trip time synchronized with the external source. This method cannot be used for continuous wave (CW) pumped passively mode-locked oscillators, which simply change their repetition rate by the change of the cavity length and their wavelength remains unaltered.

We have developed a new and simple, continuously tunable, mirror based Kerr-lens mode-locked Ti:sapphire oscillator. The design is based on a mirror pair having negative GDD and at the same time positive TOD (third order dispersion). Although, it is well known for a long time that GDD and TOD in the cavity affect the operational wavelength of the Kerr-lens mode locked oscillators, this is the first time this behavior is explicitly exploited to produce a continuously tunable Ti:sapphire laser. The realized oscillator is continuously tunable via the positive GDD of a wedge-pair and operates at the wavelength where the intracavity GDD is near zero. In the present configuration, it is tunable in the 750-795 nm range, which is the relevant range for our planned spectroscopic applications, the excitation of the 229Thorium low-energy isomeric nuclear state, using the 5th harmonic in the vacuum ultraviolet regime [16,17].

In the following, first we describe the experimental setup and then the theory behind the tuning of the operational spectrum. After presenting the measurement results, we compare them to the theory.

2. Oscillator setup

The layout of the oscillator is illustrated in Fig. 1. It is a mirror-based oscillator setup with a cavity design suggested in [8,18]. The Ti:sapphire crystal is 3.6 mm long with 80% absorption at the 532 nm wavelength of the pump laser (finesse pure 16, Laser Quantum). The two curved mirrors around the crystal have focal lengths of 50 mm. The two CM-2 are usual broadband chirped-mirrors (Layertec) with -40 fs2 GDD and low TOD. The tunability of the oscillator is given by the CM-1 special mirror-pair (10Q20UF.40, Newport) with large negative GDD and large positive TOD. In the spectral range between 750 nm and 900 nm, its TOD is about +760 fs3. Together with the BK7 wedge-pair (wedge angle of 8°), it provides a continuously tunable GDD with slightly changing TOD inside the cavity. The oscillator cavity is designed with short/long arm length of about 770/1100 mm to set the repetition rate to 80 MHz. The design offers the possibility to turn the oscillator into a frequency comb by future repetition rate and CEP stabilization. Furthermore, the tuning range of 750-795 nm is suitable to generate VUV pulses in the spectral range of 150-159 nm for future spectroscopic application.

 figure: Fig. 1.

Fig. 1. Oscillator arrangement. CM: chirped-mirror; OC: output coupler; HR: high-reflector mirror; Ti:S: Ti:sapphire crystal.

Download Full Size | PDF

3. Working principle of tunability

Earlier dispersion tuning methods, as mentioned above, rely on the fact that the cavity round-trip time or the group delay GD(ω) of the cavity is spectrally dependent, and the laser becomes tunable by changing the cavity length. Our scheme is based on the spectral dependence of the intracavity group delay dispersion GDD(ω) due to a third-order dispersion TOD:

$$GDD({\omega ,d} )= GD{D_0} + GV{D_0} \cdot d + \frac{1}{3}TO{D_0}({\omega - {\omega_0}} )$$
where the index 0 refers to the quantities at λ0 = 800 nm or ω0 = 750π THz. GDD0 is the cavity GDD without the wedge-pair, GVD0 = 44.6 fs2/mm is the group velocity dispersion of the BK7 material of the wedge-pair and d is the thickness of the glass inserted by the wedge-pair. TOD0 is the intracavity TOD mainly caused by the multiple reflections on the CM-1 mirrors. Using 12 reflections, it is in the range of 9000 fs3 and the small 151 fs3 and 32 fs3/mm contributions of the sapphire crystal and the BK7 wedge-pair respectively and the small residual TOD of the other mirrors in the cavity can be neglected. A Kerr-lens mode-locked oscillator can run at two different modes: (i) At the wavelength of maximal gain of the cavity. Here, the pulse duration (spectral width) changes by changing the intracavity GDD [8,19] and the central wavelength remains almost unaffected by the intracavity GDD and TOD. Using this mode, laser operation can be achieved by both negative [7,8] and positive [20] intracavity GDD. (ii) In this mode, the central wavelength is tunable, it is strongly affected by the intracavity GDD and TOD. At the same time, the spectral width is large and remains mainly unaffected. In this case, the effect of the Kerr-lens is stronger. According to our knowledge, this mode has not been demonstrated yet. In this mode, the pulse is short, and the wavelength is determined by realizing an intracavity
$$GDD({\omega ,d} )= GD{D_{SPM}} \approx 0.$$

In fact, the intracavity GDD should be slightly negative to compensate the effect of the self-phase modulation (SPM) of the mode-locking in the laser crystal [8] to form a stable soliton. Equation (2) can be fulfilled with suitable ω-d pairs. This second mode was realized in our scheme by altering the cavity by decreasing the distance between the two curved mirrors further away from the CW optimal position. This required the application of a larger 3.8 W pump power instead of 3.5 W to reach the threshold of the stable mode locking. At a moderate tuning range, < 10% of the central wavelength, the approximation $\lambda _0^{ - 1} - {\lambda ^{ - 1}} \approx \left( {\lambda - {\lambda _0}} \right)/\lambda _0^2$ can be applied to get the operational wavelength:

$$\lambda \approx \lambda _0^\textrm{'} + \frac{{3\lambda _0^2 \cdot GV{D_0}}}{{2\pi c \cdot TO{D_0}}}d,{\kern 7pt}\textrm{ with }{\kern 7pt}\lambda _0^\textrm{'} = {\lambda _0} + \frac{{3\lambda _0^2 \cdot \left( {GD{D_0} - GD{D_{SPM}}} \right)}}{{2\pi c \cdot TO{D_0}}},$$
which means an approximately linear dependence of the operational wavelength to the thickness d of the inserted glass of the wedge-pair.

4. Experimental results

The described oscillator was realized using 12 reflections on the CM-1 mirror-pair and inserting different amounts of BK7 glass into the cavity. The oscillator was continuously tunable within a broad spectral range, between 760 nm and 795 nm as plotted in Fig. 2(a) together with measured spectra at 4 central wavelengths. Operation at 750 nm was realized by removing the wedge-pair and the generated spectrum is also plotted in Fig. 2(a) and the corresponding measurement points in Fig. 2(b). Additional peaks are observable at the pedestals of the spectra, which appear when the phase matching between dispersive wave, solitary wave and periodic perturbation is fulfilled [8]. As can be seen in Fig. 2(a) and plotted separately in Fig. 2(b), the central wavelength of the spectra is continuously shifted near-linearly with the thickness of the inserted BK7 glass, as expected from Eq. (3), with a slope of about 4.7 nm/mm. The spectral width changed from 21 nm to 25 nm, plotted in Fig. 2(b), corresponding to a transform limited pulse duration between 43 fs and 36 fs, respectively. To check the pulse durations, autocorrelation curves were measured directly at the oscillator output at 760 nm and 795 nm central wavelengths and presented in Fig. 2(c) and Fig. 2(d). At 760 nm, the autocorrelation curve shows a transform limited 44-fs-long pulse, while at 795 nm, the pulse was near to the transform limit. The measured pulse duration was 40 fs, which was about 10% longer than the transform limit. The corresponding small chirp can be easily compensated extra-cavity. The oscillator was pumped with 3.8 W power in all measurements and the output power changed slightly from 0.50 W to 0.52 W while tuning from 760 nm to 795 nm.

 figure: Fig. 2.

Fig. 2. (a) Measured spectra tuned over the 750 nm to 795 nm spectral range. (b) The central wavelength of the spectra has linear dependence on the inserted thickness of the intracavity BK7 glass (wedge-pair). The spectral width also changed by the wavelength tuning between 21 nm and 25 nm. (c) and (d) are the measured autocorrelation curves at central wavelength of 760 nm and 795 nm, respectively.

Download Full Size | PDF

5. Discussion

As mentioned above and depicted in Fig. 2(b), the operational wavelength depends linearly on the inserted BK7 glass thickness, as described by Eq. (3). The fitted trend line yields λ’0 = 745 nm and a slope of 4.7 nm/mm. According to these values, the laser should operate at 745 nm when the wedge-pair is removed. Actually, it operates at 750 nm (or slightly below, at 748 nm) as indicated in Fig. 2(b). The reflectance of the CM-1 mirrors starts to decrease at 750 nm, which impedes the oscillator to operate much below that wavelength.

Using the values of the fitted trendline in Eq. (3), it is possible to calculate the GDD0 and the TOD0 of the CM-1 mirror-pair at λ0 = 800 nm and compare them to the manufacturer values. After correcting with the TOD of the sapphire and the ≈11 mm BK7 glass, the settings necessary for the operation at 800 nm, the calculation yields 758 fs3/bounce for the TOD, which is in good agreement of the above mentioned 760 fs3/bounce manufacturer value. GDD0 means without BK7 glass, however the GDD of the laser crystal, air, and additional chirped mirrors are considered (80 fs2 for CM-1 and 40 fs2 for CM-2 at 800 nm). Other high-reflectors and the output coupler in the cavity are zero GDD mirrors (|GDD| < 20 fs2). To get the measured 745 nm operational wavelength in Eq. (3), one must consider GDDSPM = -270 fs2 dispersion to form a stable soliton. In that case, the side-peaks in the spectra should appear +45 nm and -65 nm from the central wavelength using Eq. (4) [8]

$$\frac{{GD{D_{SPM}}}}{2}\left( {\varDelta {\omega^2} + \frac{1}{{{T^2}}}} \right) + \frac{{TO{D_0}}}{6}\varDelta {\omega ^3} ={\pm} 2\pi $$
where Δω is the distance of the side-peak from the operational frequency of the laser, furthermore T = τ/1.76, and τ is the FWHM pulse width. In the measurements, the positions of the side-peaks were +35 nm and -65 nm, very close to the expected values. The small difference is attributed to the manufacturing inaccuracy of the mirrors and mainly to their fourth and higher order dispersions and the fact that the spectra have not necessarily Lorentzian shape.

6. Summary

A continuously tunable Ti:sapphie oscillator has been developed using the intracavity GDD and the large TOD of a special mirror-pair as basic tuning principle. The oscillator cavity was built using only mirrors and a thin wedge-pair. Short pulses of 40 ± 4 fs pulses were obtained in the spectral range of operation between 750 nm and 795 nm, at a repetition rate of 80 MHz. The oscillator can be tuned to longer wavelength by adding further dispersive material to the cavity; however, it is practical to use less reflections on the mirror pair instead, if operation at longer wavelength range is required.

The demonstrated oscillator is suited for converting its wavelength by high harmonic generation using gas or solid medium, which has been demonstrated at 800 nm with extra-cavity [21,22], enhancement cavity [23,24] and intra-oscillator [25] arrangements. In that way, a tunable, short pulse laser source can be obtained in the ultraviolet spectral range by 3rd harmonic generation or in the vacuum ultraviolet spectral range between 150 nm and 159 nm by 5th harmonic generation for different spectroscopic applications. Such a spectral range in the vacuum ultraviolet is of special interest, as it corresponds to the predicted excitation wavelength of the 229Thorium isomeric nuclear state [16,17].

The required moderate 3.8 W pump power opens the way for further development in the direction of increased output power and a suitable dispersion design for obtaining shorter pulse durations and an even broader tuning range.

Funding

Österreichische Nationalstiftung für Forschung, Technologie und Entwicklung (AQUnet); European Metrology Programme for Innovation and Research (20FUN01 TSCAC); Horizon 2020 Framework Programme (856415).

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. E. J. Mayer, J. Mobius, A. Euteneuer, W. W. Ruhle, and R. Szipocs, “Ultrabroadband chirped mirrors for femtosecond lasers,” Opt. Lett. 22(8), 528–530 (1997). [CrossRef]  

2. B. Golubovic, R. R. Austin, M. K. Steiner-Shepard, M. K. Reed, S. A. Diddams, D. J. Jones, and A. G. Van Engen, “Double Gires–Tournois interferometer negative-dispersion mirrors for use in tunable mode-locked lasers,” Opt. Lett. 25(4), 275–277 (2000). [CrossRef]  

3. C. Ruppert and M. Betz, “Generation of 30 femtosecond, 900-970 nm pulses from a Ti:sapphire laser far off the gain peak,” Opt. Express 16(8), 5572–5576 (2008). [CrossRef]  

4. P. G. Antal and R. Szipocs, “Tunable, low-repetition-rate, cost-efficient femtosecond Ti:sapphire laser for nonlinear microscopy,” Appl. Phys. B 107, JTuA12 (2012). [CrossRef]  

5. J. C. E. Coyle, A. J. Kemp, J.-M. Hopkins, and A. A. Lagatsky, “Ultrafast diode-pumped Ti:sapphire laser with broad tunability,” Opt. Express 26(6), 6826–6832 (2018). [CrossRef]  

6. H. Liu, G. Wang, K. Yang, R. Kang, W. Tian, D. Zhang, J. Zhu, H. Han, and Z. Wei, “Diode-pumped Kerr-lens mode-locked Ti:sapphire laser with broad wavelength tunability,” Chinese Phys. B 28, 094213 (2019). [CrossRef]  

7. A. Stingl, C. Spielmann, F. Krausz, and R. Szipocs, “Generation of 11-fs pulses from a Ti:sapphire laser without the use of prisms,” Opt. Lett. 19, 204–206 (1994). [CrossRef]  

8. C. Spielmann, P. F. Curley, T. Brabec, and F. Krausz, “Ultrabroad-band femtosecond lasers,” IEEE J. Quantum. Electron. 30(3), 1100–1114 (1994). [CrossRef]  

9. I. D. Jung, F. X. Kartner, N. Matuschek, D. H. Sutter, F. Morier-Genoud, G. Zhang, U. Keller, V. Scheuer, M. Tilsch, and T. Tschudi, “Self-starting 6.5-fs pulses from a Ti:sapphire laser,” Opt. Lett. 22(13), 1009–1011 (1997). [CrossRef]  

10. R. Ell, U. Morgner, F. X. Kärtner, J. G. Fujimoto, E. P. Ippen, V. Scheuer, G. Angelow, T. Tschudi, M. J. Lederer, A. Boiko, and B. Luther-Davies, “Generation of 5 fs pulses and octave-spanning spectra directly from a Ti:sapphire laser,” Opt. Lett. 26(6), 373–375 (2001). [CrossRef]  

11. T. Fuji, A. Unterhuber, V. S. Yakovlev, G. Tempea, A. Stingl, F. Krausz, and W. Drexler, “Generation of smooth, ultra-broadband spectra directly from a prism-less Ti:sapphire laser,” Appl. Phys. B 77(1), 125–128 (2003). [CrossRef]  

12. J. Jiang, S. Fang, G. Chang, H. Han, H. Teng, and Z. Wei, “Megawatt peak power, octave-spanning Ti:sapphire oscillators,” Appl. Phys. Express 12(10), 102009 (2019). [CrossRef]  

13. S. A. Diddams, K. Vahala, and T. Udem, “Optical frequency combs: Coherently uniting the electromagnetic spectrum,” Science 369(6501), 267–278 (2020). [CrossRef]  

14. A. Sharp, O. Kitzler, A. Fuerbach, D. J. Spence, and D. W. Coutts, “Generation of sub-100 fs ultraviolet pulses from a Kerr-lens mode-locked Ce:LiCAF laser,” Appl. Opt. 60(27), 8316–8320 (2021). [CrossRef]  

15. S. Yamashita and M. Asano, “Wide and fast wavelength-tunable mode-locked fiber laser based on dispersion tuning,” Opt. Express 14(20), 9299–9306 (2006). [CrossRef]  

16. B. Seiferle, L. von der Wense, P. V. Bilous, I. Amersdorffer, C. Lemell, F. Libisch, S. Stellmer, T. Schumm, C. E. Dullmann, A. Palffy, and P. G. Thirolf, “Energy of the 229Th nuclear clock transition,” Nature 573(7773), 243–246 (2019). [CrossRef]  

17. T. Sikorsky, J. Geist, D. Hengstler, S. Kempf, L. Gastaldo, C. Enss, C. Mokry, J. Runke, C. E. Dullmann, P. Wobrauschek, K. Beeks, V. Rosecker, J. H. Sterba, G. Kazakov, T. Schumm, and A. Fleischmann, “Measurement of the 229Th Isomer Energy with a Magnetic Microcalorimeter,” Phys. Rev. Lett. 125(14), 142503 (2020). [CrossRef]  

18. T. Brabec, P. F. Curley, C. Spielmann, E. Wintner, and A. J. Schmidt, “Hard-aperture Kerr-lens mode locking,” J. Opt. Soc. Am. B 10(6), 1029–1034 (1993). [CrossRef]  

19. D. H. Song, S. I. Hwang, and D.-K. Ko, “Dynamics of Sub-microjoule Femtosecond Pulse Formation in a Negative Dispersion Regime,” Journal of the Korean Physical Society 61(5), 730–734 (2012). [CrossRef]  

20. A. Fernández, A. Verhoef, V. Pervak, G. Lermann, F. Krausz, and A. Apolonski, “Generation of 60-nJ sub-40-fs pulses at 70 MHz repetition rate from a Ti:sapphire chirped pulse-oscillator,” Appl. Phys. B 87(3), 395–398 (2007). [CrossRef]  

21. J. Seres, E. Seres, C. Serrat, and T. Schumm, “Non-perturbative generation of DUV/VUV harmonics from crystal surfaces at 108 MHz repetition rate,” Opt. Express 26(17), 21900–21909 (2018). [CrossRef]  

22. J. Seres, E. Seres, C. Serrat, and T. Schumm, “High harmonic generation in AlN due to out-of-surface electron orbitals,” OSA Continuum 4(1), 47–54 (2021). [CrossRef]  

23. G. Winkler, J. Fellinger, J. Seres, E. Seres, and T. Schumm, “Non-planar femtosecond enhancement cavity for VUV frequency comb applications,” Opt. Express 24(5), 5253–5262 (2016). [CrossRef]  

24. J. Seres, E. Seres, C. Serrat, E. C. Young, J. S. Speck, and T. Schumm, “All-solid-state VUV frequency comb at 160 nm using high-harmonic generation in nonlinear femtosecond enhancement cavity,” Opt. Express 27(5), 6618–6628 (2019). [CrossRef]  

25. E. Seres, J. Seres, and C. Spielmann, “Extreme ultraviolet light source based on intracavity high harmonic generation in a mode locked Ti:sapphire oscillator with 9.4 MHz repetition rate,” Opt. Express 20(6), 6185–6190 (2012). [CrossRef]  

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (2)

Fig. 1.
Fig. 1. Oscillator arrangement. CM: chirped-mirror; OC: output coupler; HR: high-reflector mirror; Ti:S: Ti:sapphire crystal.
Fig. 2.
Fig. 2. (a) Measured spectra tuned over the 750 nm to 795 nm spectral range. (b) The central wavelength of the spectra has linear dependence on the inserted thickness of the intracavity BK7 glass (wedge-pair). The spectral width also changed by the wavelength tuning between 21 nm and 25 nm. (c) and (d) are the measured autocorrelation curves at central wavelength of 760 nm and 795 nm, respectively.

Equations (4)

Equations on this page are rendered with MathJax. Learn more.

G D D ( ω , d ) = G D D 0 + G V D 0 d + 1 3 T O D 0 ( ω ω 0 )
G D D ( ω , d ) = G D D S P M 0.
λ λ 0 ' + 3 λ 0 2 G V D 0 2 π c T O D 0 d ,  with  λ 0 ' = λ 0 + 3 λ 0 2 ( G D D 0 G D D S P M ) 2 π c T O D 0 ,
G D D S P M 2 ( Δ ω 2 + 1 T 2 ) + T O D 0 6 Δ ω 3 = ± 2 π
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.