Random acts of brightness
From disordered systems comes coherent light
Like musicians in a marching band, waves of laser light move "in step" with one another. As a result, the waves spread only slightly, even over great distances.
Laser light differs from the light produced by other sources, such as electric light bulbs or the sun. The light from these sources is a confusion of many wavelengths, traveling in all directions. It is called incoherent light. But a laser produces an intense, narrow beam of light that travels in only one specific direction, much like a marching band moves in unison along a parade route. This is called coherent light.
Producing laser light, light waves of a single frequency or just a few frequencies that "march" together along a straight path, requires a gain medium — a substance that will amplify light. The gain medium must also have a precise and orderly atomic structure that will allow it to store energy and then release the energy at the right time and in the right direction.
Although lasers are technological marvels of exactness and order, physicists have speculated for years about the possibility of achieving laser light with disordered systems.
Costas Soukoulis, an Ames Laboratory senior physicist, has gone beyond speculation. He and former Iowa State University graduate student Xunya Jiang, now working at DiCon Fiberoptics, Inc., near Berkeley, Calif., have developed a theoretical model that simulates the phenomenon of random lasing, in which photons that follow random paths create a multiple-light-scattering laser.
In a regular, or periodic, laser the amplifying substance is positioned within the laser's cavity. It might be a solid, such as a crystal or a semiconductor; a gas or a mixture of gases, such as helium-neon; or a liquid dye.
All in all, the production of laser light might be thought of as a continuous round-trip journey. When excited by an outside energy source, perhaps a flashlamp or another laser, the electrons in the amplifying substance get "pumped" to a higher energy state where many will spontaneously emit incoherent light waves. The excited electrons that haven't spontaneously emitted photons of light will begin to fall to a lower energy level, the metastable state, where they "hang out" for awhile, just waiting to release their excess energy as light so they can return home to the ground state. These are the electrons that will eventually emit coherent laser light.
The release of coherent light is called stimulated emission. It takes place when electrons in the metastable state are stimulated to fall to the ground state by photons coming in from other electrons that are falling from their metastable states. When an electron falls from this stimulation, it also produces a photon of light in exactly the same direction and with exactly the same energy and phase as the stimulating photon. Then, the process repeats itself. The laser light produced is reflected back and forth between mirrors positioned in the laser's cavity, a process that amplifies the light many times. Energy gain becomes greater than energy loss, and an intense laser beam is created.
Random lasers — The big surprise
But what happens if instead of possessing an orderly arrangement of atoms that produce synchronized, focused wavelengths, the structure of the light-amplifying substance is one that scatters light quite efficiently in all directions? Can such a substance produce laser light?
Soukoulis' and Jiang's model says, yes it can. A new kind of laser can be created using photons that follow random paths. The disordered system does not impede lasing — it accounts for it. The process goes something like this: When light is shined into a substance that scatters it well, the photons get bounced in random directions. If this happens often enough, it's likely that the trajectory of the photons inside the gain medium will be extremely long and that the photons will travel many times through the same crystal grains, ricocheting from side to side as they go. Under these conditions, light can be amplified tremendously. The process is similar to the way light from an ordinary laser travels back and forth between the two mirrors in the cavity. If the electrons in the disordered gain medium get pumped to a higher energy level while traveling their long, random paths, the result could be amplification to laser light. The gain medium, whether a crystal powder or a material containing random scatterers, would, in effect, become a laser.
Explaining the experiments
Soukoulis' and Jiang's theoretical work represents some of the very first efforts to understand and describe the mechanics of random lasing. They have interpreted the investigations of experimentalists Hui Cao of Northwestern University, Ad Lagendijk of the University of Amsterdam and Val Vardeny of the University of Utah, breathing logic and reason into the laboratory observations and giving credibility to the random-lasing phenomenon.
"The first thing we wanted to do was to see if we could understand all the narrow, sharp lasing peaks the experimentalists were seeing in the output spectrums of disordered systems, such as the zinc oxide nanocrystals Cao was studying," says Soukoulis, who is also an Iowa State University physics professor. "So we tried to simulate the real experiment in the computer using a complex numerical technique called finite difference time domain. We did a one-dimensional version of a three-dimensional experiment."
Jiang adds, "The advantage of the FDTD method is that from our computer simulation we can see the dynamic process of the random system and how the electric field is building up inside. We are able to follow the evolution of the electric field."
The ability to track the population of the electrons as they are pumped to higher energy levels and then fall to metastable and ground energy states has allowed Soukoulis and Jiang to arrive at a number of important results.
Using their computer simulation of random lasing, the two theorists are able to determine the threshold value of lasing for the amplifying substance. Knowing the threshold value is critical because it is the amount of energy needed to pump the electrons in the amplifying substance to the intensity where lasing takes place. "One very important thing we discovered is that the threshold value of lasing decreases as disorder increases," says Soukoulis. "So, the more the concentration of scatterers, the lower the threshold of lasing."
Through their one-dimensional model, Soukoulis and Jiang also found that the sharp peaks the experimentalists had observed in the output spectrums were coming from specific lasing modes of the disordered systems. "From our theory, we can really solve and understand these beautiful peaks," says Soukoulis. He explains that for disordered systems the lasing has a greater probability of occurring in some particular regions of the system than in others. As the electrons in the amplifying random substance are pumped to a higher energy level, the lightwaves retain their shape, but they become bigger and bigger. Eventually they will start lasing.
In addition, the model revealed that if the pumping intensity is increased above a maximum value, the number of lasing modes remains the same. They do not increase anymore, rather; they saturate to a constant value that is determined by the degree of randomness in the system and its length.
Predictions and possibilities
"With our one-dimensional model, we can predict exactly where the lasing modes will be in a given random system, which mode will lase best, and the wavelength, which determines the color of the emission light," says Soukoulis.
The ability to predict the lasing modes in a given random system is critical when considering some of the potential uses for random lasers. The novel devices hold promising properties for brightening the pixels in flat-panel displays. By reducing the size of the phosphor grains that emit light in these displays, it might be possible for the electron emitter in each pixel to excite the phosphor's electrons and initiate lasing, which would brighten the pixels.
Another possible application for random lasers lies within the medical arena. Used with chemical sensors, these devices could provide a sophisticated, noninvasive tool for diagnosing problems within the human body. And paint-on random lasers may one day light the way to more efficient and economical search-and-rescue missions to identify downed ships and airplanes, and even individual passengers.
For more information:
Costas Soukoulis, (515) 294-2816
Research funded by:
DOE Office of Basic Energy Sciences