Chaos comes to light in asymmetric microlasers, making them a thousand times more powerful


In the June 5, 1998 cover story of "Science", researchers from the Max Planck Institute for the Physics of Complex Systems in Dresden/Germany, Yale University in New Haven, Connecticut, and Lucent Technologies' Bell Laboratories in Murray Hill, New Jersey, report on an innovative laser design which relies on the presence of chaotic light rays inside a resonator with a cross section close to the width of a human hair.

At roughly 0.05 millimeter diameter, the tiny cylinders made of semiconductor material are among the smallest ever made, belonging to a class of microlasers that have been pioneered by Sam McCall, Richard Slusher and coworkers at Bell Labs in the early Nineties. Already, larger semiconductor lasers are at the heart of numerous everyday items, such as CD players. The key difference between these conventional devices and the microlaser lies in the shape: to create the perfectly synchronized photons that make laser light so special, a resonator has to be formed by trapping the light. The early pioneering microresonators consisted of perfectly round disks that can store light in wavefronts circulating around the rim - squeezed toward the edge like the passengers of a runaway merry-go-round, but nonetheless caught for great lengths of time.

What holds back the light is total internal reflection, the same effect that makes the surface of a calm swimming pool look like a shiny mirror to a diver submerged in the water. The new laser design still uses this effect because it is the key to a small resonator volume, but traps the light in a bowtie-shaped pattern which is unlike anything that can be achieved with round resonators. The very existence of this unconventional lasing pattern is a consequence of chaos, induced by the highly oval cross section with which the semiconductor microcylinders have now been fabricated by Bell Labs scientists Claire Gmachl, Federico Capasso, Deborah Sivco, Alfred Cho and Jérôme Faist (who is now with the Université de Neuchâtel, Switzerland).

Capasso, head of the Semiconductor Physics Research department, and Faist had earlier invented the quantum-cascade laser, a fundamentally new type of laser that operates like an electronic waterfall. Multiple layers Gallium-Indium-Arsenide and Aluminum-Indium-Arsenide are stacked on top of each other with atomic precision in such a way that mid-infrared light is generated when an electric current passes through this semiconductor sandwich. The recent breakthrough came when this technology was combined with the notion of asymmetric resonant cavities (see here for further pictures and information) introduced by physicist Jens Nöckel, now with the Max Planck Institute, and Douglas Stone, chairman and professor of Applied Physics at Yale.

One thing which round, i.e. symmetric, microlasers had been unable to deliver, was emission in the form of focused beams. Several years ago Stone, Nöckel and Yale coworker, Professor Richard K. Chang discovered that a suitably deformed oval resonator shows highly focused laser emission in certain directions, whereas round resonators radiate equally all around their perimeter. The oval shape allows trapped light rays to move on irregular - chaotic - paths after several reflections at the surface, similar to the unpredictable paths of billiard balls after a few collisions among each other. Why chaos, of all things, should bring about focused emission, can be understood by analysing the hidden structure underlying the ray chaos, which is not random, but deterministic. Essentially, the merry-go-round theory remains valid to a certain extent, except for some very unpleasant wobbling that causes a slingshot effect in specific directions. Initial experimental and theoretical results confirming the idea were published in Optics Letters, Physical Review Letters and Nature.

The Bell Labs experiment, however, goes far beyond merely confirming Nöckel's and Stone's previous theoretical ideas. The enormous increase in output power, accompanied by emission in four narrow, controllable beams, is observed when the oval deformation of the semiconductor cylinder is pushed to an extent at which no significant lasing can come from circulating wavefronts bearing any resemblance to those in a round resonator. Uncovering the deterministic structure in the ray paths for these parameters, Nöckel, Stone and Yale applied physicist Evgenii Narimanov found that the bowtie pattern (see here for further pictures and information) emerges as the sole candidate for trapping the laser light. It preserves a degree of regularity that is crucial for focused emission.

In these experimental lasers, basic research in quantum-chaos physics has been combined in an interdisciplinary way with technological innovation at the forefront of the highly competetitive quest for new laser sources: While the systematic shape design of the experimental Bell Labs lasers was guided by theoretical predictions, the simultaneous theoretical exploration was in turn led into a new direction in direct response to the experimental progress. What has thus been learned about chaotic microresonators can now again help guide the development of future devices. As this research has shown, the challenge will be to embrace chaos as a helpful tool, aiming to find the proper balance between chaos and regularity.

Asymmetric microlasers have left the stage of trial and error in which their development threatened to become stalled while no appropriate theory existed. Because of their low power consumption and small size, they can be packed onto chips for optical computing, and offer great promise for the growing fiber-optic telecommunication industry. The mid-infrared lasers with which the chaotic-resonator has been demonstrated and characterized are themselves potentially useful in pollution monitoring, breath analysis in medicine and other applications. However, the same concepts can be applied to a broad range of solid-state laser materials emitting in the visible part of the spectrum or at shorter infrared wavelengths relevant for fiber communications, thereby moving us further into the age of photonics.

Published: 5-06-98
Jens U. Nöckel
Max Planck Institute for the Physics of Complex Systems

Phone: +49-351-871-2208
Fax: +49-