New Ytterbium Host Offers Efficiency and Tunability

Breck Hitz

The broad absorption and emission spectra of ytterbium, together with its small quantum defect, make it an ideal laser dopant — in many ways, it is superior to the quintessential veteran, neodymium. The chief drawback of ytterbium is the small splitting between its terminal laser level and the ground state, which results in a nonnegligible thermal population of its terminal laser level. This thermal population causes reabsorption losses, increased threshold and diminished overall efficiency.

Experiments published earlier this year reported a new ytterbium host, Gd₂SiO₅ (GSO), with a much larger splitting — 1067 cm^–1 — between the terminal laser level and the ground state. Although these studies established the favorable laser qualities of Yb:GSO, the crystal material proved to be delicate and readily subject to cleavage.

To overcome this problem, scientists at the Chinese Academy of Sciences in Shanghai and Beijing have grown an alloyed host crystal, Gd_2(1–x)Y_2xSiO₅ (GYSO), by combining GSO and Y_2xSiO₅ (YSO). Their investigation of the new material has led them to conclude that it could be one of the best hosts yet developed for ytterbium lasers.

Figure 1. A fiber-coupled diode laser pumped the experimental Yb:GYSO laser. The 976-nm pump radiation exited the resonator through the folding mirror, which was reflective at the laser wavelength but transmissive for the pump. Either the simple output coupler or the prism and output coupler (shown in the box) were incorporated into the resonator. Images ©OSA.

The investigators grew their Yb:GYSO (x = 0.5) crystal by the Czochralski technique from a 50/50 melt of GSO and YSO, with 5 percent atomic ytterbium impurity introduced into the melt. The resulting 70-mm-long, 35-mm-diameter boule contained 5 percent atomic ytterbium and had excellent physical properties, with no tendency to cleave. Spectroscopically, the splitting between the terminal laser level and the ground state — 995 cm^–1 — was somewhat less than in Yb:GSO, but still greater than that of other ytterbium hosts. The spontaneous lifetime of the upper laser level was 1.92 ms.

Figure 2. Input vs. output data indicated that optimum coupling for the experimental laser was ~12 percent.

A three-mirror, folded resonator matched the pump-light mode to the TEM₀₀ laser mode in their experimental arrangement (Figure 1). The researchers wrapped a polished but uncoated 5 × 5 × 3-mm piece of Yb:GYSO in indium foil and mounted it on a water-cooled copper block inside the resonator. By temperature-tuning the wavelength of the fiber-coupled diode pump laser to the absorption peak of the Yb:GYSO, they achieved 75 percent absorption of the incident pump power in the crystal.

By experimenting with different output couplers, the investigators determined that optimum coupling for the simple, three-mirror resonator was ~12 percent (Figure 2). Under optimum coupling conditions, they observed a maximum output of 2.44 W at 1081.5 nm from an absorbed pump power of 5.54 W, with a slope efficiency of 57 percent.

Figure 3. With an SF10 prism in the resonator, the researchers tuned the Yb:GYSO laser smoothly from 1030 to 1089 nm.

To evaluate the wavelength tunability of the Yb:GYSO laser, the researchers inserted an SF10 prism into the resonator and observed a relatively smooth tuning curve from 1030 to 1089 nm (Figure 3). The wide, smooth tuning range bodes well for the possibility of stable, femtosecond mode-locked Yb:GYSO lasers because ultrashort pulses require a wide spectral bandwidth.

Optics Express, April 17, 2006, pp. 3333-3338.