Color center lasers have long provided an alternative to dye lasers as sources of tunable radiation, but their commercial viability has diminished in recent years as competing, tunable solid-state sources have improved. Recently, researchers from the Russian Academy of Sciences in Moscow and from IPG Photonics Corp. in Oxford, Mass., reported what they believe is the first instance of a color center laser pumped with a continuous-wave diode laser. Because the diode laser’s energy efficiency is much better than that of older pump lasers (e.g., Nd:YAG and Nd:YLF), and because its wavelength is more suitable for pumping the color center than that of older pump lasers, the development could lead to significant improvements in commercial color center lasers.The scientists studied the LiF:F–2 color center crystal, an efficient, tunable source of coherent infrared radiation when pumped with pulsed, solid-state lasers at the ~1-μm wavelength. In fact, because the performance of the LiF:F–2 laser is improved considerably by shortening the pump wavelength from 1064 nm (Nd:YAG) to 1047 nm (Nd:YLF), they had good reason to anticipate even better performance at the diode wavelength of 970 nm.They used a 5-mm-thick LiF:F–2 crystal with plane-parallel antireflection-coated end facets, mounted on a water-cooled copper heat sink. The 970-nm fiber-coupled diode pump laser provided up to 12 W of power, nearly half of which was absorbed in a single pass through the crystal. The color center laser’s resonator had a flat dichroic mirror through which the pump light entered and a concave output coupler whose reflectivity varied from 99.5 to 90 percent over the 1145- to 1160-nm spectral range, where the researchers expected the laser’s peak output to be. This nonoptimal coupling would prove to be a limiting factor on laser performance.Figure 1. The output of the color center laser varied significantly from one crystal to another, presumably a result of the crystals’ differing losses at the laser wavelength. Images ©OSA.The results varied significantly from one LiF:F–2 crystal to another because of the crystals’ differing losses at the laser wavelength (Figure 1). The best results, about 46 μJ of output with a slope efficiency of 14 percent relative to absorbed pump power, were obtained with 80-μs pump pulses at a 5-kHz repetition frequency. When the scientists operated the pump laser continuously, they observed an ∼50 percent drop in efficiency. The laser’s linewidth at maximum pump power was ∼12 nm, centered at 1149 nm. They believe that a better output coupler — i.e., one with high reflectivity (∼0.99) across a wider spectral range — would have increased both the magnitude and the linewidth of the laser’s output.The laser exhibited abnormal behavior as a function of temperature. Although most lasers operate better at lower temperatures, the output of the color center laser increased with rising temperature (Figure 2).Figure 2. Unlike most lasers, the output of the color center laser increased with rising temperature.The investigators explained this phenomenon as an effect of intersystem crossings and absorption into the quadruplet electronic states of the LiF:F–2 color centers. The output of the color center laser was highly sensitive to intracavity losses, so even a small population in the quadruplet levels diminished the laser’s output. This was balanced, however, by the fact that intersystem crossings become faster at higher temperatures and quickly depopulate the quadruplet levels.Thus, the intracavity loss due to quadruplet absorption decreased — at least initially — as the temperature of the laser crystal increased. This can be seen in the relation among the curves in Figure 2 (i.e., higher output at higher heat-sink temperature at any constant repetition frequency), and in particular in the shape of the 35 °C curve: Above several kilohertz, crystal heating by the pump light increased its temperature, and the output increased with increasing temperature.Figure 3. The unusual effect of temperature was apparent in the steady-state operation of the laser. An elevated temperature was required for continuous-wave operation.Temperature-related effectsFurther evidence of the unusual behavior appeared in the steady-state operation of the laser (Figure 3). When the pump power was switched on, the laser output dropped quickly from its initial value, presumably because of a rapid increase in the population of the quadruplet levels. When the heat-sink temperature was 10 °C, the laser never recovered. The population of the quadruplet levels remained high, and the resulting intracavity loss kept the laser below threshold. With a heat-sink temperature of 25 °C, the initial power again dropped as the population of the quadruplet levels built, but then the elevated temperature allowed rapid intersystem crossings, so the quadruplet population diminished and the output reached an equilibrium value. At an even higher temperature, the quadruplet population dropped quickly, but other high-temperature effects — reduced luminescence lifetime and ground-state absorption — overwhelmed laser oscillation and eventually extinguished the laser.Optics Letters, July 15, 2006, pp.