Stabilized QC lasers sense in mid-infrared

Researchers at the Pacific Northwest National Laboratory (PNL; Richland, WA) and at Lucent Technologies (Murray Hill, NJ) have achieved relative linewidths of 5.6 Hz between two quasi-independent cavity-locked quantum-cascade (QC) lasers and subsequently used them to perform sub-Doppler spectroscopy at 8.5 μm.

Laser stabilization is fundamental to research and development of ultrasensitive cavity-enhanced chemical sensors, said Matthew Taubman of PNL, who presented the research results in May at the annual Conference of Lasers and Electro-Optics (CLEO; Long Beach, CA). The researchers were seeking to demonstrate the extension of the stabilization technology from the visible to near-infrared regions down to the mid- to long-wave infrared, in particular, using the QC laser technology developed by Lucent.

Reducing linewidths

To achieve a 5.6-Hz (full width at half-maximum) beat between the two laser systems, the researchers had to achieve a reduction on the order of 25,000 in the free-running linewidth of the QC lasers of about 160 kHz. The accomplishment required reduction to a very high level of all but the slowest components of the frequency noise contributing to these lasers' linewidths, Taubman said. He added that the slowest components of the noise were removed artificially to allow the measurement to be taken.

The QC lasers were frequency stabilized to separate optical cavities using highly optimized servo loops based on the Pound-Dreyer-Hall technique (see figure). Unlike traditional stabilization experiments in which both lasers are stabilized to adjacent modes of a single cavity under stringent engineering constraints, the two separate cavities in the PNL experiment were made of stainless steel vacuum fittings and rubber O-rings bolted directly to an optical table with minimal vibration isolation.

Despite the fact that the cavities exhibited "significant acoustic resonances," the experiment was successful, Taubman said, because of a third servo loop of lower gain and bandwidth. The third loop could not modify the fast linewidths of the laser systems, but effectively removed relative drift and low frequency noise, forcing one laser to track the other at low audio-band frequencies and enabling a highly stable 5.6-Hz relative linewidth beat measurement.

"This is important because in this state, such stabilized lasers represent a sharpened spectroscopic tool," Taubman said. "With the capacity for such narrow linewidths, extremely narrow spectroscopic features can be observed. This is of great interest to the chemical detection community." The technique may also represent a new way of measuring the performance of stabilized lasers, he added. Although not an entirely independent measurement, it allows a precision comparison to be made without the need for extensive acoustic, thermal, and barometric isolation of optical cavities.

Taubman also reported successful development, by the same research group using similar stabilization technology, of a chemical sensor using a technique developed at the University of Colorado by Ye, Ma and Hall, known as NiceOhms (noise-immune cavity-enhanced optical heterodyne molecular spectroscopy).¹ They reported a sensitivity to optical absorption of 2.9 x 10^-12/cm for observation times of the order of 1 s, compared to the world record of 1 x 10^-14/cm. "These techniques will continue to be refined and engineered to specific applications, with the end goal of developing deployable sensors based on this and other techniques using QC lasers," Taubman said.

REFERENCE
1. J. Ye, L.-S. MA, J. J. Hall, Opt. Lett. 21, 1000 (1996)