Microchip Laser Exploits Yb:LuAG Properties
Researchers create passively Q-switched Yb:LuAG laser.
Ytterbium-doped lutetium aluminum garnet (Yb:LuAG) has several desirable qualities as a solid-state laser material. Compared with the more familiar Yb:YAG, it has a larger effective emission peak cross section and a higher thermal conductivity. It absorbs more efficiently at 970 nm, allowing higher quantum efficiency than can be obtained with 940-nm pumping, and spectroscopic studies of the material have indicated that it should be well suited for ultrashort-pulse generation. Now scientists at the University of Electro-Communications in Tokyo and at the Russian Academy of Sciences in Moscow have demonstrated what they believe is the first passively Q-switched microlaser using Yb:LuAG and have generated pulses as short as 610 ps.
Figure 1. Two 8-mm focal-length lenses focused the pump light to an ∼100-μm spot in the Yb:LuAG laser crystal. (OC = output coupler.) Images reprinted with permission of Optics Letters.
The scientists were constrained, however, by the equipment available to use a 940-nm, fiber-coupled diode laser as the pump. A pair of lenses focused the pump light into a 1.12-mm-long Yb:LuAG crystal whose entrance face was coated for high transmission at the pump wavelength and for high reflectivity at the 1030-nm lasing wavelength (Figure 1). The crystal was doped with 8.2 atomic percent ytterbium, and its other face was antireflection-coated at the laser wavelength. The total resonator length, from the 80 percent reflecting output coupler to the high-reflecting coating on the Yb:LuAG, was only 1.62 mm. Tightly sandwiched between the Yb:LuAG crystal and the output mirror was a 0.5-mm-long uncoated piece of Cr4+:YAG, which served as the saturable absorber to Q-switch the laser.
The scientists characterized the laser with no saturable absorber in the cavity and then switched between two Cr4+:YAG crystals to study its Q-switched performance. The first absorber had a 90 percent transmission before saturating, which increased to 97.5 percent when saturated; the second had unsaturated and saturated transmissions of 95 percent and 99 percent, respectively. (The scientists measured the saturated transmissions by bleaching the crystals with an intense pulse of 1064-nm light.)
Figure 2. There was no rollover at the high end of the input-output transfer function, indicating that the laser would be capable of higher output power with a more powerful pump. (T0 = initial transmissions, ηs = slope efficiency).
Not surprisingly, the average power output was greatest with no saturable absorber in the resonator, and the absorber with the higher absorption reduced the average power even more (Figure 2). There is a trade-off here, however, because the high-absorption Q-switch produced shorter pulses — typically between 600 and 700 ps, compared with pulse durations in excess of a nanosecond with the low-absorption Q-switch. In practice, the scientists say, the Q-switch would be selected to optimize the laser’s performance in a particular application.
Figure 3. The 12.8-kHz pulse train was very stable, with less than 5 percent pulse-to-pulse amplitude and timing jitter (a). Individual pulses were as short as 610 ps, measured as shown (b).
The short — 1.62 mm — resonator generated widely spaced longitudinal modes, and etalon effects arising from the uncoated Cr4+:YAG crystal provided a powerful mode-selection mechanism. Nonetheless, the gain bandwidth of Yb:LuAG was sufficient to allow several longitudinal modes, spaced by three times the cavity’s free spectral range, to oscillate simultaneously.
The laser’s pulse repetition frequency increased linearly with absorbed pump power. At 12.8 kHz, with the high-absorption Q-switch in the cavity, the laser generated a very stable 12.8-kHz train of 610-ps pulses (Figure 3).
Optics Letters, Nov. 15, 2007, pp 3266-3268.
MORE FROM PHOTONICS MEDIA