The intrinsic limits of quantum cascade lasers

One of the hallmarks of lasing is a dramatic narrowing of the light’s frequency spread. In 1958 Arthur Schawlow and Charles Townes deduced that the laser linewidth is fundamentally limited by unavoidable spontaneous emission. (Thanks to other sources of noise, a real laser’s linewidth is usually considerably broader.) Semiconductor diode lasers required a revision of the intrinsic linewidth formula to account for additional inherent broadening, but quantum cascade lasers (described in Physics Today, May 2002, page 34 ) had been thought to obey the original limit. Now Saverio Bartalini and colleagues at Italy’s National Institute of Optics-CNR, the European Laboratory for Non-linear Spectroscopy, and the Second University of Naples have confirmed a recent theory predicting that QCLs can in fact beat the Schawlow-Townes limit and yield significantly improved spectral purity. Key to the 2008 theory by Masamichi Yamanishi and coworkers at Hamamatsu Photonics was the recognition that nonradiative transitions in QCLs strongly suppress spontaneous emission. To test the prediction, the Italian researchers tuned their IR QCL to be halfway down a carbon dioxide absorption peak at 4.33 µm (69.3 THz). Thanks to the steep slope of the absorption curve there, frequency fluctuations were converted into detectable intensity variations. That technique enabled the team to measure the noise spectrum over seven decades of frequency and to extract the intrinsic QCL linewidths for various pump currents. The obtained widths, in the range of 500 Hz, agreed well with the new theory and were three orders of magnitude smaller than predicted by the venerable Schawlow-Townes formula. (S. Bartalini et al., Phys. Rev. Lett. 104 , 083904, 2010 http://dx.doi.org/10.1103/PhysRevLett.104.083904 .)