A high power LiF:F2- color center laser is demonstrated with broadband emission. The excitation source is a quasi-continuous wave diode side-pumped acousto-optically Q-switched Nd:YAG laser. Under an incident 1064-nm laser power of 25.4 W, the highest output power of up to 4.7 W is obtained with a macro pulse repetition rate of 400 Hz and a micro pulse repetition rate of 50 kHz. The broadband emission is centered at 1142 nm with a bandwidth of 13 nm.
© 2015 Optical Society of America
LiF:F2- color center lasers (CCL) have been capable to generate tunable laser lines ranged from 1100 nm to 1300 nm [1–3]. This band has been difficult to access through rare earth doped laser gain materials. LiF:F2- crystal displays high thermal stability even at room temperature and can be irradiated by high intensity radiation without color center destruction . It has a high luminescence quantum efficiency (η = 0.65) at room temperature , and large absorption and emission cross-sections (~7 × 10−17 cm2) . In addition, wide absorption band (0.8-1.1 µm) allows to use commercially available neodymium or ytterbium doped solid state lasers as pumping sources. So CCL based on LiF:F2- crystals have attracted much attention [1–13]. In [6–8], broadband oscillations were obtained with different pumping schemes. Especially, high pulse energy of up to 100 J was obtained by using high pulse energy Nd-glass laser as the excitation source . Widely tunable lasers were realized in [1, 3, 5]. Single longitudinal mode operations were reported in  and . A linewidth of less than 100 MHz was obtained in . Through frequency doubling of this kind of tunable lasers, researchers were able to generate tunable visible laser lines [10,11]. Laser amplifications were also realized by using pico-or nano-second pulses [12,13]. In most of the published results, the excitation sources were flash-lamp pumped lasers, which limited the overall conversion efficiency.
In recent years, we have paid attention to LiF:F2- CCLs excited by diode-pumped lasers in order to increase the conversion efficiency. In 2011, using an diode-pumped 1064-nm Nd:YVO4 laser as the pump source, we investigated the broadband and tunable characteristics of LiF:F2- CCL . For the broadband operation, an output power of 230 mW was reached under 1.6 W of pumping power. For the tunable laser operation, tuning range from 1.10 µm to 1.29 µm was obtained by using Littrow grating scheme. In 2014, we reported further results on tunable LiF:F2- CCL . The pumping source was a similar diode-pumped Nd:YVO4 laser at 1064 nm. By designing Littman grating cavity, we realized a tuning range cavity, we obtained single longitudinal mode operation at 1178 nm. These results showed great potential of LiF:F2- CCLs excited by diode- pumped lasers. However, output power was limited by the low power of the pumping lasers.
In this paper, we report a high power broadband LiF:F2- CCL pumped by a diode side-pumped 1064 nm Nd:YAG laser. The Quasi-continuous-wave (QCW) diodes were controlled to generate macro pulses with repetition rates of 300 to 400 Hz and pulse duration of 300 µs. An acousto-optical (AO) Q-switch was employed to generate micro pulses with repetition rate of 50 kHz. Under an incident 1064-nm laser power of 25.4 W, the highest output power of up to 4.7 W was obtained with a macro pulse repetition rate of 400 Hz. All the powers referred to in this paper are power averaged during all the time. To the best of our knowledge, this is the highest output power of LiF:F2- CCL. The broadband emission is centered at 1142 nm with a bandwidth of 13 nm. The beam quality factors (M2) were determined to be 4.7 and 4.3 for horizontal and vertical directions, respectively.
2. Experimental setup
The experimental setup for the high power broadband LiF:F2- CCL is shown in Fig. 1. The excitation source was a QCW diode-side-pumped acousto-optically Q-switched Nd:YAG laser emitting at 1064 nm. The diodes could generate macro pulses with repetition rate of 1500 Hz and pulse duration of 100-300 µs. The dimension of the Nd:YAG crystal rod was Ø3 mm × 70 mm and the doping concentration was 1 at.%. Both surfaces were anti-reflection (AR) coated at 1064 nm (R<0.2%). A 46-mm-long acousto-optic (AO) Q-switch with AR coatings (R<0.2%) on both faces at 1064 nm was employed to generate micro pulses, with a pulse repetition rate of 50 kHz. The rear mirror (M1) was a 3000 mm radius-of-curvature concave mirror with high reflectivity (HR) coating at 1064 nm (R>99.8%). The output coupler (OC) M2 was a plane mirror coated for partial reflectivity (R = 75%) at 1064 nm. In order to obtain linearly-polarized 1064 nm radiation, a Brewster polarizer (BP) was inserted in the cavity. The λ/4 plate was used to compensate the thermally induced depolarization loss . All optical elements were placed as close as possible. The total cavity length was 100 mm. The Faraday isolator (ISO) was used to avoid the radiation reflected back into the 1064 nm laser cavity. The polarization plane of the 1064 nm radiation was horizontal, which resulted in a minimal reflected loss on the surface of the Brewster-cut LiF:F2- crystal. The λ/2 plate was employed to adjust the power transmitted from the isolator. A convex lens (L) with a focal length of 300 mm was used to enhance the intensity of the 1064 nm laser. The Brewster-cut LiF:F2- CC crystal was 38-mm-long with an aperture of 8 × 20 mm2. The initial absorption coefficient at 1064nm of the CC crystal was 0.2 cm−1. The input mirror (M3) for the CCL cavity was coated for high transmission (HT) at 1064 nm (T >99.5%) and HR in 1100-1200 nm spectral region (R>99%). The plane mirror M4 coated for partial reflection (PR) at 1142 nm was used as the output coupler. A dichroic mirror (M5) coated for HR at 1064 nm (R >99.8%) and HT at 1100-1200 nm (T>99.5%) was used to separate the CC laser from 1064 nm laser. The overall cavity length was 70 mm. The Nd:YAG crystal, AO Q-switch and LiF:F2- crystal were all water cooled. The water temperature was maintained at 20 °C.
3. Results and discussions
To begin with, we studied the output characteristics of the 1064 nm laser. The output powers are shown in Fig. 2.
With a Q-switched pulse repetition rate (i.e. micro pulse repetition rate) of 50.0 kHz, we studied the output power of 1064 nm laser with different macro pulse repetition rate at the pulse duration of 300 µs. The results are shown in Fig. 2(a). The output powers were measured by an EPM2000 power meter and a PM30 detector. At the macro repetition rate of 400 Hz, we obtained the highest output power of 27.6 W with a pumping current of 75 A (corresponding to diode power of 180 W). The output powers for 300 Hz and 350 Hz were 24.0 W and 20.5 W, respectively. The highest output power of 25.0 W for 300 Hz was obtained with the highest pumping current of 85 A. It is necessary to notice that during all the experiments below, the macro pulse duration was kept at 300 µs and micro pulse repetition rate was kept at 50.0 kHz.
Figure 2(b) shows the micro pulse widths with respect to the pumping current at different macro pulse repetition rates. The time characteristics were monitored with a digital oscilloscope (Tektronix DPO 4104B) and a fast photodiode (Thorlabs, DET10A/M). For different macro pulse repetition rates, the lowest pulse widths occurred at the pumping current of 65 A. With pumping current increasing further, the pulse widths increased. This phenomenon could be addressed to multi-mode operations. When pumping current was higher than 65 A, the laser began to run at multi-mode scheme. The higher-order modes led to pulse width increasing. Multi-mode operation could also result in the output power increasing. This can be found Fig. 2(a), where the output power dramatically increased with pumping current higher than 65 A. With output power higher than 20 W, the pulse durations ranged from 65 ns to 80 ns. These values are comparable with the fluorescence time of 55 ns of LiF:F2- crystal . So this laser was suitable to excite the LiF:F2- laser.
As Fig. 2(a) shows, when output power higher than 20 W was concerned, the laser operating at 400 Hz needed the lowest pumping level to generate the same output power. When the pumping current was higher than 71 A, the output 1064 nm laser power became unstable. As a result, we selected 71 A as the working current in the following experiments. Under the pumping current of 71 A, we obtained the maximum output power of 25.4 W at the macro pulse repetition rate of 400 Hz. The corresponding conversion efficiency, electrical to 1064 nm optical radiation of the diode side-pump Nd:YAG laser was 6.9%. The output powers were 19.0 W and 21.5 W at 300 Hz and 350 Hz, respectively. The pulse durations for 300, 350 and 400 Hz was 65, 70 and 70 ns, respectively.
We began the CCL experiments by choosing a 1000 mm radius-of-curvature concave mirror as the input mirror (M3). We put the crystal in the focus of convex lens (L) and the pump beam waist radius was 400 µm. The output coupler was coated for PR at 1142 nm (R = 39%). The output power of the broadband CCL is shown in Fig. 3.
By fixing the pumping current at 71 A, the vertical axis in Fig. 3 was obtained by rotating the λ/2 plate ahead of the isolator and the power was measured behind the convex lens (L). At the macro pulse repetition rate of 400 Hz, we found the oscillating threshold for the CCL was below 2.9 W. With increasing the pumping power, the output power for CCL kept increasing. The maximum output power reached 4.7 W at the pumping power of 25.4 W, corresponding to an optical-to-optical conversion efficiency of 18.5%. This efficiency was higher than that we obtained in . At the same pumping current of 71 A, we also tried the CCL at the macro pulse repetition rates of 300 Hz and 350 Hz. The experimental results were also shown in Fig. 3. The output power was obtained to be 3.9 W and 4.1 W at 300 Hz and 350 Hz, respectively.
Although we got such high output power with a large pump power, we didn’t observe any degradation of F2- CC in our crystal. The experiments by Dr. Ilichev showed that the photodestruction of F2- CC from the excited state could have a two-photon photoionization mechanism and this could happen when LiF:F2- CC crystal operated as a passive Q-switch and pumped with a very high intensity pump radiation . In our experiment the LiF:F2- CC crystal operated as an active element of the tunable laser and the pump radiation was continuously absorbed and converted into a tunable radiation in the crystal. In additional the degradation of F2- CC depends on the concentration of other types of CC (F2, F3-, colloidal) which depends on the technology of the preparation of F2- CC. Our LiF:F2- sample had the reduced concentration of F3- and colloidal CC . So we found that there was no degradation of F2- CC in our experiments.
The optical spectrum was measured with an optical spectrum analyzer (Yokogawa AQ6370C, 600-1700 nm). A typical result is shown in Fig. 4. The center wavelength was measured to be 1142 nm with a spectral width of 13 nm. It was recorded at the highest output power of 4.7 W.
Then under the macro pulse repetition rate of 400 Hz, we tested another two input mirrors with different curvature radii of 500 mm, and 2000 mm. The measured output powers are shown in Fig. 5(a). The highest output powers of 4.1 W and 2.63 W were obtained with curvature radii of 500 mm and 2000 mm, respectively. It was found that the input mirror with curvature radius of 1000 mm was most appropriate to this CCL.
By using the input mirror with a curvature radius of 1000 mm, we tried another two output couplers with reflectivity of 55% and 35%. The output powers are shown in Fig. 5(b). At the same pumping power, the output powers with different output couplers were a little higher than the other two cases.
Under the macro repetition rate of 400 Hz, input mirror with a curvature radius of 1000 mm and output coupler with reflectivity of 39%, we also studied temporal and spatial characteristics of the CCL.
By using the same digital oscilloscope (Tektronix DPO 4104B) and fast photodiode (Thorlabs, DET10A/M), we measured the pulse duration of the CCL. The results are shown in Fig. 6.
With the pumping power increasing from 3.0 W to 25.4 W the pulse duration of the CCL increased from 9.0 ns to 62.7 ns. This phenomenon could be explained by analyzing typical pulse shapes of the pumping laser and the CCL presented in Fig. 7. When the incident 1064 nm laser power was low, only one single pulse of CCL oscillated and its pulse duration was as short as 9.0 ns, as shown in Fig. 7(a). While, when the incident 1064 nm laser power was high enough, besides the sharp pulse, the remaining pump radiation could excite another big pulse, as shown in Fig. 7(b). From Fig. 7(b), it could also be found that the overall CCL pulse shape and pulse duration were quite similar to the pumping laser. When pump pulse power was high enough, the CCL operated in a quasi-CW regime and its pulse behaved as a slave to the pumping pulse, which was reasonable.
The beam quality factor M 2 measurements were performed by focusing the beam of the CCL into a CCD camera using a biconvex lens (f = 200 mm). By measuring the laser spot radius and fitting the experimental results, we determined the M 2 of the CCL at the maximum output power. As shown in Fig. 8, the beam quality factors in vertical and horizontal directions were 4.7 and 4.3, respectively.
We have demonstrated that high power LiF:F2- color center laser excited by a high power acousto-optically Q-switched QCW diode side-pumped Nd:YAG laser. At the Q-switched repetition rate of 50.0 kHz, we studied different cavity mirrors of the CCL with different macro repetition rate. The optimal result occurred at the macro repetition rate of 400 Hz with the input mirror of a curvature radius of 1000 mm and the output coupler of a reflectivity of 39%. The highest output power of up to 4.7 W was obtained under the incident 1064 nm laser power of 25.4 W, corresponding to an optical-to-optical conversion efficiency of 18.5%. The broadband emission was centered at 1142 nm with a bandwidth of 13 nm. The beam quality factors in vertical and horizontal directions were determined to be 4.7 and 4.3, respectively. The conversion efficiency of the whole laser system (electrical to 1142 nm optical radiation) was about 1.3% that is much higher than that with flash lamp pumped laser system.
This work was supported by the National Natural Science Foundation of China (Contract no. 61378032 and 1141101110), Shanghai Key Laboratory Program of All Solid-state Laser and Applied Techniques (ADL-2014004), and the Russian Foundation for Basic Research (grant no. 15-52-53026).
References and links
1. T. T. Basiev, P. G. Zverev, A. G. Papashvili, and V. V. Fedorov, “Temporal and spectral characteristics of a tunable LiF: F2- colour-centre crystal laser,” Quantum Electron. 27(7), 574–578 (1997). [CrossRef]
2. J. C. Diettrich, I. T. McKinnie, D. M. Warrington, and V. V. Ter-Mikirtychev, “Tunable, single axial mode LiF:F2- laser,” Opt. Commun. 204(1-6), 317–322 (2002). [CrossRef]
4. W. Gellermann, A. Mutter, S. Wilk, and F. Luty, “Formation, optical properties, and laser operation of F2- centers in LiF,” J. Appl. Phys. Lett. 61, 1297–1303 (1987).
5. T. T. Basiev, A. Y. Dergachev, A. Y. Karasik, V. V. Fedorov, and R. L. Shubochkin, “Highly efficient generation of tunable picosecond pulses in an LiF:F2- laser crystal,” Quantum Electron. 26(12), 1042–1044 (1996). [CrossRef]
6. T. T. Basiev, S. V. Vassiliev, V. A. Konjushkin, and V. P. Gapontsev, “Pulsed and cw laser oscillations in LiF:F-2 color center crystal under laser diode pumping,” Opt. Lett. 31(14), 2154–2156 (2006). [CrossRef] [PubMed]
7. T. T. Basiev, B. V. Ershov, S. B. Kravtsov, S. B. Mirov, V. A. Spiridonov, and V. B. Fedorov, “Lithium fluoride color center laser with an output energy of 100 J,” Sov. J. Quantum Electron. 15(6), 745–746 (1985). [CrossRef]
8. V. V. Fedorov, P. G. Zverev, and T. T. Basiev, “Broadband lasing and nonlinear conversion of radiation from LiF:F2+ and LiF:F2- colour centre lasers,” Quantum Electron. 31(4), 285–289 (2001). [CrossRef]
9. T. T. Basiev, A. G. Papashvilli, V. V. Fedorov, S. V. Vassiliev, and W. Gellermann, “Single-Longitudinal-Mode Pulsed LiF:F2- Color-Center Laser for High Resolution Spectroscopy,” Laser Phys. 14(1), 23–29 (2004).
10. S. M. Giffin, I. T. Mckinnie, and V. V. Ter-Mikirtychev, “Tunable yellow-green laser based on second harmonic generation of LiF:F2- in KTP,” Opt. Quantum Electron. 31(1), 35–41 (1999). [CrossRef]
11. S. M. Giffin, G. W. Baxter, I. T. McKinnie, and V. V. Ter-Mikirtychev, “Efficient 550-600-nm tunable laser based on noncritically phase-matched frequency doubling of room-temperature LiF:F2- in lithium triborate,” Appl. Opt. 41(21), 4331–4335 (2002). [CrossRef] [PubMed]
12. T. T. Basiev, S. V. Garnov, V. I. Vovchenko, A. Ya. Karasik, S. M. Klimentov, V. A. Konyushkin, S. B. Kravtsov, A. A. Malyutin, A. G. Papashvili, P. A. Pivovarov, and D. S. Chunaev, “Direct amplification of picosecond pulses in F2-:LiF crystals,” Quantum Electron. 36(7), 609–611 (2006). [CrossRef]
13. T. T. Basiev, A. Ya. Karasik, V. A. Konyushkin, V. V. Osiko, A. G. Papashvili, and D. S. Chunaev, “Amplification of picosecond pulses in F2-:LiF crystals synchronously pumped by picoseconds and nanosecond laser pulses,” Quantum Electron. 35(4), 344–346 (2005). [CrossRef]
14. P. G. Zverev, Z. J. Liu, X. Y. Zhang, and V. A. Konyushkin, “High repetition rate LiF: F2- color center laser,” Laser Phys. 21(9), 1549–1553 (2011). [CrossRef]
15. S. J. Men, Z. J. Liu, Z. H. Cong, P. G. Zverev, V. A. Konushkin, X. Y. Zhang, S. S. Zhang, and Y. Liu, “Tunable narrow line-width LiF: F2- color center laser,” Opt. Commun. 324, 160–164 (2014). [CrossRef]
16. R. Kandasamy, M. Yamanaka, Y. Izawa, and S. Nakai, “Analysis of Birefringence compensation using a quarter-wave in solid-state lasers,” Opt. Rev. 7(2), 149–151 (2000). [CrossRef]
17. N. N. Ilichev, A. V. Kir’yanov, A. A. Malyutin, P. P. Pashinin, and S. M. Shpuga, “Bleaching of the F2- color centers in an LiF crystal due to two-photon absorption from the excited state,” Sov. Phys. JETP 71(3), 532–537 (1990).