Abstract

Future lightwave systems are envisioned where the optical waves are treated like present day radio waves, allowing angle modulation, heterodyne detection, and many multiple channels via frequency multiplexing. One essential device for such systems is the analog of the crystal oscillator to provide an absolute frequency reference in the optical domain. Most previous work on absolute stabilization of semiconductor lasers has been done in the 830-nm region.¹ In the low loss regions of 1.5 (Ref. 2) and 1.3 μm,³ experiments using the generally more complex and faint molecular spectra have required long absorption cells due to low absorption coefficients. Circumventing the lack of useful atomic transitions originating in the ground state, the optogalvanic effect⁴-⁵ has been used to stabilize the frequency of DFB lasers at 1.3 and 1.5 μm. The technique using the optogalvanic signal is of particular interest due to its simplicity, compactness, wide choice of reference lines within the telecommunications window, and high ratio of signal strength to laser power. These signals are strong and well separated and thus facilitate their use in lightwave systems. However, the optogalvanic effect is still not well understood. For example, the optogalvanic signal observed from the famous He-Ne laser transition at 632 run is ~2 orders of magnitude smaller than that horn typical transitions. This suggests that each atomic line of interest must be examined separately to evaluate its applicability to laser stabilization. We surveyed twenty-three excited state transitions of argon, krypton, and neon around 1.3 and 1.5 μm. All exhibit the optogalvanic effect, but there is significant variation in the magnitude of the effect.

© 1990 Optical Society of America