Lasers in the mid-infrared spectral region — often described as the wavelength range between 3 and 30 μm — are very useful in molecular spectroscopy because many molecules have unique spectral signatures in that region. With sufficient power, such lasers are also very attractive as optical pumps for lasers operating at even longer wavelengths and as laser-surgical scalpels with extremely high (micron scale) resolution. In addition, in the noncommercial world, mid-infrared lasers are extremely important for photonic countermeasures because heat-seeking ordnance can be effectively disabled by such lasers.All these considerations underlie the significance of the recent demonstration at the University of New Mexico in Albuquerque of what researchers there believe is the most powerful fiber laser yet developed that is based on ZBLAN, a fiber material composed of zirconium, barium, lanthanum, aluminum and sodium combined with fluorine. Although its nearly 10 W of power is generated at 2.78 μm — a tad short of the commonly defined mid-infrared region — the laser is nonetheless of interest for all the applications mentioned above.Unlike conventional silica-glass optical fibers, ZBLAN-glass fibers transmit well in the infrared region but at the price of reduced strength, thermal stability and corrosion resistance. Low-concentration erbium-doped ZBLAN isn’t an ideal laser material because the lower laser level of the ~2.8-μm transition is long-lived: Its spontaneous lifetime is 9 ms, compared with a 6.9-ms lifetime for the upper laser level. That means that ions tend to pile up quickly in the lower level until their population matches that of the upper level, defeating the population inversion.Figure 1. Erbium-doped ZBLAN fiber is not an ideal laser material because the lower laser level (4I13/2) has a long (9 ms) spontaneous lifetime. There are two ways to work around this issue. First, the fiber can be co-doped with praseodymium, in which case the 4I13/2 lower laser level can be depopulated by transferring its energy to the praseodymium 3F3 level (arrow ET1 above). Alternatively, if the erbium doping is high enough, an up-conversion process can transfer energy from the 4I13/2 lower laser level to the 4I7/2 and 4I15/2 levels (arrows ETU1 above). Parallel processes (ET2 and ETU2) depopulate the upper laser level, but are weaker than the mechanisms that depopulate the lower level. Images reprinted with permission of Optics Letters.There are two work-arounds to obtain continuous-wave lasing in Er:ZBLAN (Figure 1), and the researchers chose to work with erbium-only doped fibers containing a high concentration of dopant. They used the Fresnel reflection from one end of the ZBLAN fiber and an external highly reflecting mirror to form their laser resonator (Figure 2). The fiber had a 15-μm-diameter core that was surrounded by a rectangular 200 × 250-μm inner cladding, which was, in turn, surrounded by a 370-μm-diameter outer cladding. The core was doped with 6 molecular percent erbium, and the fiber was fabricated by FiberLabs of Saitama, Japan. The researchers pumped this fiber with a 100-W, 975-nm collimated laser diode from Apollo Instruments of Irvine, Calif.Figure 2. The fiber-laser resonator was formed between the Fresnel reflection at one end of the fiber and an external mirror at the other end. A 100-W, 975-nm diode array pumped the fiber’s inner cladding, and a dichroic mirror oriented at 45° to the optical axis separated the output laser power from the incoming pump light.After launching 42.8 W of pump power into the fiber, the researchers observed 9 W of output at 2785 nm. At lower pump powers, the output was relatively constant, but at higher pump powers, they saw severe (10 to 20 percent) fluctuations in the output, which they attribute to nonlinear effects and excited-state absorption in the ZBLAN fiber. Power density at the peaks of these fluctuations approached the fiber’s damage threshold. By fixing the pump end of the fiber in a water-cooled metal housing, the scientists diminished, but did not eliminate, these fluctuations.The key to even higher power from Er:ZBLAN lasers may lie in optimizing the erbium concentration. The current result, when compared with earlier research, indicates that increasing the dopant concentration to 6 percent results in higher outputs. The researchers speculate that there is likely to be an optimal doping level — probably even greater than 6 percent — and that further investigation is needed.Optics Letters, Jan. 1, 2007, pp. 26-28.