QCL experts invent Raman laser chip

07 Mar 2005

The first electrically-pumped Raman laser is reported in a paper in Nature.

A team of US scientists has made the world’s first electrically-driven Raman laser. The development could lead to a new type of semiconductor laser that operates at wavelengths that conventional laser chips cannot reach.

Raman chip

Although lasers that rely on Raman scattering (a nonlinear optical effect that amplifies light signals) have been demonstrated before, to date they have always required a powerful external laser to pump them.

Federico Capasso, one of the inventors of the quantum cascade laser (QCL), and his co-workers from Harvard University, Texas A&M University and Bell Labs, have now come up with an ingenious solution that gets around the problem.

“Our lasers combine the advantages of nonlinear optical devices and of semiconductor injection lasers,” report the team in the 24 February issue of Nature (Nature 423 845). “They could lead to a new class of compact and wavelength-agile mid and far-infrared light sources.”

The design exploits Raman scattering between quantum wells within the active region of a QCL to produce a very compact semiconductor Raman laser. When electrically-pumped, the QCL starts to lase and its emission at 6.7 microns is Raman shifted to 9 microns.

“The laser itself is microscopic -- about 10 microns wide, 6 microns thick and 2 mm long,” Mariano Troccoli, from the Harvard University group told Optics.org. “When packaged with contacts, thermoelectric cooler and DC power source it can fit into a box of 2 inches in size.”

By carefully designing the structure of their InGaAs/InAlAs QCL the team says that they can achieve highly efficient Raman conversion of around 30%, which is orders of magnitude stronger than that usually observed in silica optical fibres.

To date, the team has made ten samples and achieved Raman lasing in all of them. Its prototypes emit just under 20 mW at a drive current of 4 A and operate at temperatures of up to 170 K. The researchers now plan to increase the temperature of operation and demonstrate lasing at other wavelengths.

“In a different material system such as InAs/AlSb, lasing at telecom wavelengths could be possible,” commented Troccoli. “It would also be very interesting to obtain Raman lasing at very long wavelengths in the terahertz region. It seems to be feasible but there is a lot of work to be done.”

Author
Oliver Graydon is editor of Optics.org and Opto & Laser Europe magazine.