Scientists at Yale University claim to have developed an "anti-laser." The device, created by a Yale team including Dr. Douglas Stone, takes the laser concept in reverse: In a normal laser, the beam is created by feeding light or electricity through a gain medium like gallium arsenide, with reflectors positioned to keep the beams bouncing through. As the light bounces back and forth, the medium adds more photons to the mix (which is why it's called a "gain" medium), and one of the reflectors is partially transparent to let the amplified beam through—the laser.
In Stone's anti-laser device, the structure is similar, with two important differences: the incoming laser is countered with a beam that's the opposite of itself, and the medium, silicon, is optimized to be make the beam experience a loss of coherence rather than a gain. The result is that the two beams dissipate in the medium and the energy is released as heat.
While Stone feels his discovery, called a coherent power absorber, is important, and he has patented the technique, he cautions against calling it the "world's first" anti-laser device.
"There are devices sort of like this that have already been produced," Stone told PCMag. "We have new kind of principle. What they are looking for is things that are small and that don't use much power."
The "they" Stone is talking about is anyone in the field of "high-performance" computing and the integration of optical technologies with traditional electrical semiconductors found in computer chips. IBM, HP, Motorola, and Intel are all in the field of "photonics," Stone says. The overall design of a chip with this technology puts a semiconductor layer below an optical layer, with an "interface" layer in between.
"Roughly speaking, the reason there's so much interest in this, is that if you make circuits insanely small—as they are now—they crosstalk and the circuit doesn't work the way it's supposed to, which is why they're heading to this design."
Stone's creation could be a significant development in the optical layer, since its light-dissipating abilities are exactly what's needed in these systems.
"For all theses devices, you need to manipulate light. It needs to be done small and for cheap. The basic components you need are filters that affect certain wavelengths and not others; modulators, which take an incoming light beam and reduce its intensity; and detectors or transducers that take in the photons and output electrical energy.
"Our device can definitely do all three functions," he says.
While the development shows promise, Stone says it's just a proof-of-concept for now, and that the current roadmap for computers that use photonic technology estimates that high-performance supercomputers (like IBM's Watson) will use it by 2015, with personal computers possibly getting the tech by 2020.
For Stone's invention to make it into new computing technology, there are significant engineering hurdles to overcome. However, he says he's already able to overcome one: that the laser energy is all dissipated as heat.
"This experiment didn't collect the energy, but that's easy to do," he says. "What you need to do is apply a voltage across your device, so you can sweep it out and collect the electrons. In principle, our device can be very small, and very soon we'll have a version that doesn't generate heat but generates electricity."
