Investigators at Aculight Corp. in Bothell, Wash., have employed noncritical phase matching to obtain 236-nm radiation by twice frequency-doubling a Q-switched, 946-nm Nd:YAG laser. They generated 43 mW of average ultraviolet power inside a cesium lithium borate (CLBO) crystal from 190 mW of blue power, for a second-harmonic conversion efficiency of 22.6 percent.In general phase matching, called critical phase matching, the momentum vectors of the ordinary, fundamental wave and of the extraordinary, second-harmonic wave propagate through a nonlinear crystal in a direction such that both waves experience the same refractive index. The extraordinary wave in a birefringent crystal, however, "walks off" from the ordinary wave unless both are traveling perpendicular to or parallel to the optic axis. This walk-off, and a related phase mismatch caused by the sharply focused beams, can be very detrimental to second-harmonic conversion efficiency.Propagating both beams perpendicular to the optic axis is known as noncritical phase matching. It is not always possible, but in some crystals and at some wavelengths, the crystal's natural indices are close enough to the proper values that noncritical phase matching can be accomplished by tuning the crystal's temperature, a method the Aculight researchers used with the CLBO.By cooling the crystal to 15 °C, the group obtained noncritical phase matching between the 473-nm fundamental beam (itself produced by frequency-doubling the 946-nm output of an Nd:YAG laser in periodically poled KTP) and the 236-nm second-harmonic beam. Although 43 mW of ultraviolet radiation was generated inside the crystal, Fresnel and other losses reduced the measurable In the setup, a fiber-coupled diode laser pumped the 946-nm Nd:YAG laser with up to 25 W. The 2.7 W (when pumped with 18 W) of output of the Nd:YAG was obtained in 95-ns pulses with a measured M2 of 1.01. The KTP crystal produced 100 mW of stable 473-nm output, or 200 mW accompanied by some power loss and crystal degradation.The researchers focused the 200-mW blue beam onto the 10-mm, cooled CLBO. Because the phase matching was noncritical, the conversion efficiency to 236 nm was 20 times greater than it would have been with a similar, beta-barium borate crystal.They believe that shorter wavelengths can be generated efficiently in CLBO by mixing the 236-nm photons with longer-wavelength photons. Mixing the 236-nm photons with readily available 1.08-µm photons, for example, could create a solid-state competitor to the 193-nm ArF laser.Several practical difficulties must be addressed before such a system would be commercially feasible, however. Maintaining the crystal stably at –15 °C is more costly than not controlling crystal temperature. Moreover, CLBO is hydroscopic, so it must be sealed in an environment free of water vapor. In their experiment, the scientists locked the crystal in a vacuum-tight, nitrogen-filled container with orthogonal Brewster windows for the input and output beams.