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

Passively mode-locked Yb:Lu2O3 laser

Open Access Open Access

Abstract

Given its specific thermal characteristics, the sesquioxide crystal Lu2O3 is a particularly promising laser host material. We demonstrate mode locking of a Yb:Lu2O3 laser by use of a semiconductor saturable-absorber mirror. The laser emits up to 470 mW in the picosecond regime, corresponding to a pump efficiency as high as 32%. With dispersion compensation, pulses as short as 220 fs at an average power of 266 mW are obtained at 1033.5 nm. To our knowledge, this is the first demonstration of a femtosecond oscillator based on Yb:Lu2O3.

©2004 Optical Society of America

1. Introduction

Ytterbium doped materials are attractive for compact high-power femtosecond laser sources in the 1-µm spectral range. Femtosecond laser operation has been successfully demonstrated in a great number of Yb-doped hosts [15], and mode-locked output powers in the Watt range were achieved [6]. A particular attraction of Yb-doped materials is their small quantum defect, given the small Stokes shift (~600 cm-1) between absorption and emission, which reduces the thermal load and increases the laser efficiency. Additionally, excited state absorption, up-conversion processes and cross-relaxation are practically absent in Yb-doped materials. Moreover, the broad absorption bands of these laser materials are covered by high-power InGaAs laser diodes, which permits efficient pumping.

The emission and absorption cross sections, the upper state lifetime, the splitting of the two electronic states of Yb3+ and the phonon energies depend strongly on the host and its anisotropy. The host also determines the thermo-mechanical properties and the allowable doping level. In this regard, the isotropic sesquioxides Sc2O3, Y2O3 and Lu2O3 [7] are promising for high-power applications because their thermal conductivity is higher and the ground state splitting is larger than in YAG. Furthermore, the Yb-doped sesquioxide crystals exhibit broader emission bands than Yb:YAG which makes them an excellent choice for highly efficient ultrashort-pulse laser sources. In quasi-three-level laser materials, a large ground-state splitting is important for a reduction of the thermal population of the lower laser level, resulting in less reabsorption losses. These advantages of Yb-doped sesquioxide crystals have already been exploited in cw laser experiments. Diode pumping of these materials in simple two-mirror cavities resulted in remarkable slope efficiencies [8]. Because of their very promising features, Yb-doped sesquioxides are also in the focus of current research on ceramic based laser materials [9].

Very recently, the first mode-locked laser operation based on a sesquioxide host was demonstrated in Yb:Sc2O3 [10]. Highly efficient mode-locked laser operation with pulses as short as 230 fs at an average power of 0.54 W was obtained at 1044 nm. Femtosecond operation was also demonstrated in a Yb:Y2O3 ceramic [11]. Here a pulse duration of 615 fs and an average power of 420 mW were achieved. The emission was centered at 1076 nm, which corresponds to the long wavelength band of the Yb-doped sesquioxide emission characteristics (see Fig. 1). Both lasers were passively mode-locked by a saturable absorber mirror (SAM), and diode-pumped operation was demonstrated.

Here we analyze the specific thermal characteristics and demonstrate, for the first time to our knowledge, mode-locked laser operation of Yb3+:Lu2O3. We present results obtained both, in the picosecond and the femtosecond regime.

2. Yb-doped sesquioxides

2.1 Thermal conductivity

The thermal conductivity of solids is a key parameter in electronics and optics, given the present tendency of increasing the energy density in functional materials. The thermal conductivity in insulator crystals is governed by lattice vibrations. Local heating of the crystal leads to population of phononic states which dissipate through the crystal lattice. The heat conductivity is limited by phonon scattering at the sample surfaces, surface defects, lattice defects, and impurities, as well as umklapp-processes depending on the sample temperature [12]. The thermal conductivity of a crystalline material usually decreases when doped by luminescent ions.

The thermal conductivity of yttria (Y2O3) was first measured by Klein and Croft [13] as 27 W/m/K. However, we could not reproduce such high values in our crystal samples. The thermal conductivity measured in our sesquioxide crystals, undoped and with Yb-doping are compiled in Table 1 and compared with corresponding values of YAG and Yb:YAG. Without Yb-doping, all investigated sesquioxide crystals generally show a higher thermal conductivity than YAG. Among the sesquioxides scandia (Sc2O3) has the highest thermal conductivity of 16.5 W/m/K, lutetia (Lu2O3) with 12.5 W/m/K the lowest at room temperature. To investigate the dependence of the thermal conductivity on Yb-doping, sesquioxide crystals with an identical doping level of 2.7–2.8% were chosen. Due to the higher cation density in sesquioxides, this doping level is comparable to 5% Yb-ion densities in YAG. Inspection of the measured values in Table 1 reveals that ytterbium-doping has the strongest effect in Sc2O3, with a reduction of the thermal conductivity to less than half of the value for the undoped host. A strong reduction is also observed in Y2O3. Despite the higher thermal conductivity of the undoped sesquioxides, neither Sc2O3 nor Y2O3 sustain this advantage when doped with Yb; both materials do not exhibit significantly improved thermal conductivities compared to the 6.8 W/m/K of Yb:YAG [14].

Tables Icon

Table 1. Thermal conductivity in W/m/K of different sesquioxide crystals in comparison to YAG with and without Yb-doping. Values in [] are estimated.

In contrast to the other sesquioxides, however, Lu2O3 stands out with only a soft decrease of the thermal conductivity to 11.0 W/m/K due to ytterbium doping. This can be explained by the fact that ytterbium and lutetium exhibit very similar masses and bonding forces. Therefore, ytterbium-doping causes only a small disturbance to the crystal lattice and affects phonon scattering only slightly. Consequently, the thermal conductivity of Yb-doped lutetia is nearly twice higher than that of Yb-doped YAG. This renders lutetia a very promising host for ytterbium based high power laser systems. Our experimental result confirms theoretical predictions of Gaumé et al. [15], who already proposed Lu2O3 as the material with the lowest drop in thermal conductivity when doped with ytterbium ions.

2.2. Spectroscopic properties

Comparative studies of 13 available Yb-doped hosts evaluating the output yield (Pout/Ppump) and the slope efficiency on the basis of the spectroscopic characteristics predicted that KGd(WO4)2, KY(WO4)2, Sc2O3, and Lu2O3, are the most efficient representatives of this class of materials for longitudinal pumping [16]. The monoclinic double tungstates KGd(WO4)2 and KY(WO4)2 stand out because of their large absorption and emission cross sections [3]. In contrast, the isotropic sesquioxides Sc2O3 and Lu2O3 [8] are more promising for high-power applications as the strong anisotropy of the thermo-mechanical properties of KGd(WO4)2 and KY(WO4)2 is a serious limitation.

The absorption and emission cross sections of Yb:Lu2O3 are shown in Fig. 1. The laser wavelength near 1033 nm corresponds to the transition from the lowest level of the 2F5/2 upper state to the third level of the 2F7/2 lower state. This transition exhibits the highest gain resulting in a reduced reabsorption loss (inset Fig. 1).

 figure: Fig. 1.

Fig. 1. Absorption σabs and emission cross section σem of Yb:Lu2O3 and energy-level diagram with the relevant transitions (inset). The points indicate the pump- and laser transitions.

Download Full Size | PDF

3. Experimental laser setup

The laser experiments are performed with a Ti:sapphire laser as a pump source, which emits up to 2 W at 976 nm. The cavity setup is illustrated in Fig. 2. For the experiments a 1.33-mm-thick 2.7% Yb/Lu-site (7×1020 Yb3+ atoms/cm3) Lu2O3 crystal was employed with an aperture of 3×5 mm2 and used under Brewster angle. No special provision was made for cooling the sample. The Ti:sapphire beam was focused by an f=6.28 cm lens through the folding mirror M2, and 70% of the incident pump radiation were absorbed by the crystal.

 figure: Fig. 2.

Fig. 2. Setup of the mode-locked Yb: Lu2O3 laser: SAM - saturable absorber mirror; M1 - focusing mirror (ROC=10 to 15 cm); M2, M3 - folding mirrors (ROC=10 cm), P1, P2 - SF6 prisms; M4, M5 - output couplers (T=1 to 5%).

Download Full Size | PDF

We studied a Z-shaped astigmatically compensated resonator with two curved folding mirrors (radius-of-curvature: ROC=10 cm) with a 30-µm cavity waist at the position of the Yb:Lu2O3 crystal. One arm contained an additional focusing mirror M1 (ROC=10 or 15 cm) to increase the intensity on the saturable absorber mirror, which acts as an end mirror. The other arm contained a plane output coupling mirror. In this arm two dispersion compensating prisms could be included. The SAM (BATOP GmbH, Germany) high-reflection band extends from 995 to 1095 nm with 2% of saturable absorption, 70 µJ/cm2 saturation fluence, and a 20 ps relaxation time constant. The non-saturable loss was specified by the supplier as less than 0.3%.

4. Yb:Lu2O3 mode-locked laser: results

Using the setup of Fig. 2 with a 5% output coupler, yet without intracavity prisms, the laser operated in the picosecond regime at a pulse repetition rate of 109 MHz. In the mode-locked regime the maximum pump efficiency reached 32%. Pulses as short as 1.16 ps at 1033.5 nm could be achieved at a maximum average output power of 470 mW. The measured slope efficiency amounted to 44%. Although the Yb:Lu2O3 crystal was not actively cooled, thermal problems did not occur. The measured autocorrelation traces are well fitted assuming a sech2-pulse shape. In Fig. 3 the autocorrelation functions are shown together with the emission spectra at maximum output power. Both, experimental data (squares) and the fit (line) are plotted. The 5.5 nm broad spectrum (FWHM) could support pulses of about 200 fs duration, which means that the experimentally obtained pulse duration in the picosecond regime exceeds the Fourier limit by a factor of 6 (Fig. 3(a)). At lower output powers the pulse duration remained unchanged but the pulse spectral width decreased by 70%.

 figure: Fig. 3.

Fig. 3. Autocorrelation trace and spectrum (inset) in the picosecond (a) and in the femtosecond regime (b).

Download Full Size | PDF

For femtosecond operation, the ROC=15 cm focusing mirror M1 was used to increase the pulse fluence on the SAM. Additionally, two 60°-SF6 prisms with a tip-to-tip separation of 41 cm were inserted in the output coupling arm (Fig. 2). These prisms compensate for the positive group-velocity dispersion inside the cavity and balance the self-phase modulation introduced by the Kerr nonlinearity of the laser crystal. For femtosecond operation, a 3% output coupler was used. The resulting pulse repetition rate is 97 MHz. The deconvolved FWHM of the pulse is 220 fs, and the central wavelength is 1033.5 nm (Fig. 3(b)). The time-bandwidth-product amounts to 0.334, which is close to the Fourier limit (0.315). The obtained pulse duration is substantially shorter than the 340 fs limit reported for Yb:YAG [2], which can be attributed to the larger gain bandwidth of Yb:Lu2O3. In contrast to the ceramic-based results of Ref. [11], stable short-pulse operation was achieved in the band with the largest emission cross section (apart from the zero-line band at 976 nm), which also coincides with the most pronounced Yb:YAG transition at 1032 nm.

The output power versus the absorbed pump power in the femtosecond configuration is shown in Fig. 4, both below and above the mode-locking threshold. The measured cw lasing threshold amounted to 380 mW of absorbed pump power. The slight increase of output power in the transition from the cw to the mode-locked regime is due to the saturated loss of the SAM. Mode-locked operation was obtained at a maximum output power of 266 mW, corresponding to a pulse energy of 3 nJ or 14-kW peak power. From these experimental data we deduce a slope efficiency of 23% and a maximum pump efficiency of 15.5%. The lower efficiency compared to the picosecond results is mainly due to insertion losses of the SF6 prisms. The observed transversal mode structure of the Yb:Lu2O3 laser remained basically TEM00. For all arrangements investigated, the mode-locked operation was stable for hours, without evidence of dropping out or tendencies toward passive Q-switching [17].

 figure: Fig. 4.

Fig. 4. Output power versus absorbed pump power of the femtosecond Yb:Lu2O3 laser and below the mode-locking threshold (cw – continuous wave).

Download Full Size | PDF

Compared to Yb:Sc2O3 [10], Yb:Lu2O3 exhibits a somewhat lower efficiency. This can be attributed to the imperfect crystal quality of the available Yb:Lu2O3 sample. The stronger visible fluorescence in lutetia is an indication for existing impurities in the crystal lattice.

5. Summary

In conclusion, we have generated the shortest pulses out of a sesquioxide laser and experimentally demonstrated that Yb:Lu2O3 is a very promising isotropic crystal for high-power mode-locked lasers in the 1 µm range. Compared to Yb:YAG, lutetia exhibits a larger gain bandwidth and higher thermal conductivity. In contrast to other Yb-doped hosts, Yb doping does not significantly corrupt the thermal conductivity in Lu2O3.

Self-starting passive mode-locking with pulses as short as 220 fs was achieved at 1033.5 nm. The demonstrated maximum output powers of 470 mW and 266 mW in the picosecond and femtosecond regime, respectively, are not yet a limit. Output powers in the Watt range can be expected when crystals with a higher optical quality – less impurities – and higher pump powers are applied.

References and links

1. F. Brunner, G. J. Spühler, J. Aus der Au, L. Krainer, F. Morier-Genoud, R. Paschotta, N. Lichtenstein, S. Weiss, C. Harder, A. A. Lagatsky, A. Abdolvand, N. V. Kuleshov, and U. Keller, “Diode-pumped femtosecond Yb:KGd(WO4)2 laser with 1.1-W average power,” Opt. Lett. 25, 1119 (2000). [CrossRef]  

2. C. Hönninger, R. Paschotta, M. Graf, F. Morier-Genoud, G. Zhang, M. Moser, S. Biswal, J. Nees, A. Braun, G. A. Mourou, I. Johannsen, A. Giesen, W. Seeber, and U. Keller; “Ultrafast ytterbium-doped bulk lasers and laser amplifiers;” Appl. Phys. B 69, 3 (1999). [CrossRef]  

3. P. Klopp, V. Petrov, U. Griebner, and G. Erbert, “Passively mode-locked Yb:KYW laser pumped by a tapered diode laser,” Opt. Express 10, 108 (2002), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-10-2-108 [CrossRef]   [PubMed]  

4. F. Druon, S. Chénais, P. Raybaut, F. Balembois, P. Georges, R. Gaumé, P. H. Haumesser, B. Viana, D. Vivien, S. Dhellemmes, V. Ortiz, and C. Larat, “Apatite-structure crystal, Yb3+:SrY4(SiO4)3O, for the development of diode-pumped femtosecond lasers,” Opt. Lett. 27, 1914 (2002). [CrossRef]  

5. M. J. Lederer, M. Hildebrandt, V. Z. Kolev, B. Luther-Davies, B. Taylor, J. Dawes, P. Dekker, J. Piper, H. H. Tan, and C. Jagadish, “Passive mode locking of a self-frequency-doubling Yb:YAl3(BO3)4 laser,” Opt. Lett. 27, 436 (2002). [CrossRef]  

6. E. Innerhofer, T. Südmeyer, F. Brunner, R. Häring, A. Aschwanden, R. Paschotta, C. Hönninger, M. Kumkar, and U. Keller, “60-W average power in 810-fs pulses from a thin-disk Yb:YAG laser,” Opt. Lett. 28, 367 (2003). [CrossRef]   [PubMed]  

7. K. Petermann, G. Huber, L. Fornasiero, S. Kuch, E. Mix, V. Peters, and S. A. Basun, “Rare-earth-doped sesquioxides,” J. Lum. 87–89, 973 (2000). [CrossRef]  

8. K. Petermann, L. Fornasiero, E. Mix, and V. Peters, “High melting sesquioxides: crystal growth, spectroscopy, and laser experiments,” Opt. Mat. 19, 67 (2002). [CrossRef]  

9. J. Kong, J. Lu, K. Takaichi, T. Uematsu, K. Ueda, D. Y. Tang, D. Y. Shen, H. Yagi, T. Yanagitani, and A. A. Kaminskii, “Diode-pumped Yb:Y2O3 ceramic laser,” Appl. Phys. Lett. 82, 2556 (2003). [CrossRef]  

10. P. Klopp, U. Griebner, V. Petrov, K. Petermann, and V. Peters, in Advanced Solid-State Photonics, OSA Technical Digest, (Optical Society of America, Washington DC, 2003), p. 154–157, P. Klopp, V. Petrov, U. Griebner, K. Petermann, V. Peters, and G. Erbert, “Highly-efficient mode-locked Yb:Sc2O3 laser,” Opt. Lett.29, 391 (2004).

11. A. Shirakawa, H. Yagi, J-F. Bisson, J. Lu, M. Musha, K. Ueda, T. Yanagitani, T. S. Petrov, and A. A. Kaminski, “Diode-pumped mode-locked Yb3+:Y2O3 ceramic laser”, Opt. Express 11, 2911 (2003), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-11-22-2911. [CrossRef]   [PubMed]  

12. N. W. Ashcroft, I. Mermin, and N. David, in Solid State Physics, Hartcourt Brace College Publishers (1976).

13. P. H. Klein and W. J. Croft, “Thermal conductivity, Diffusivity, and Expansion of Y2O3, Y3Al5O12, and LaF3 in the Range 77°–300°K,” J. Appl. Phys. 38, 1603 (1967). [CrossRef]  

14. K. Contag, S. Erhard, and A. Giesen, “Calculation of optimum design parameters for Yb:YAG thin disk lasers,” OSA Trends Opt. Photonics 34, 124 (2001).

15. R. Gaumé, B. Viana, D. Vivien, J.-P. Roger, and D. Fournier, “A simple model for the prediction of thermal conductivity in pure and doped insulating crystals,” Appl. Phys. Lett. 83, 1355 (2003). [CrossRef]  

16. G. Boulon, A. Brenier, L. Laversenne, Y. Guyot, C. Goutaudier, M.-T. Cohen-Adad, G. Metrat, and N. Muhlstein, “Search for optimized trivalent ytterbium doped-inorganic crystals for laser applications,” J. Alloys and Compounds 341, 2 (2002). [CrossRef]  

17. C. Hönninger, R. Paschotta, F. Morier-Genoud, M. Moser, and U. Keller, “Q-switching stability limits of continuous-wave passive mode locking,” J. Opt. Soc. Am. B 16, 46 (1999). [CrossRef]  

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 (4)

Fig. 1.
Fig. 1. Absorption σabs and emission cross section σem of Yb:Lu2O3 and energy-level diagram with the relevant transitions (inset). The points indicate the pump- and laser transitions.
Fig. 2.
Fig. 2. Setup of the mode-locked Yb: Lu2O3 laser: SAM - saturable absorber mirror; M1 - focusing mirror (ROC=10 to 15 cm); M2, M3 - folding mirrors (ROC=10 cm), P1, P2 - SF6 prisms; M4, M5 - output couplers (T=1 to 5%).
Fig. 3.
Fig. 3. Autocorrelation trace and spectrum (inset) in the picosecond (a) and in the femtosecond regime (b).
Fig. 4.
Fig. 4. Output power versus absorbed pump power of the femtosecond Yb:Lu2O3 laser and below the mode-locking threshold (cw – continuous wave).

Tables (1)

Tables Icon

Table 1. Thermal conductivity in W/m/K of different sesquioxide crystals in comparison to YAG with and without Yb-doping. Values in [] are estimated.

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.