Light switch promises powerful computers

By Kimberly Patch, Technology Research News

At first glance, a switch is a simple concept. It is either on or off.

Today's computer chips harbor millions of microscopic electrical switches. These transistors turn on when an electromagnetic field generated by a control electrode lowers the transistor's resistance to the flow of electrons, which allows electrical current to flow from one end of the device to the other. The presence or absence of this flow represents a 1 or a 0 of digital computing.

Circuits that switch light rather than electricity would make for faster computers, but it's difficult to use a beam of light to turn another light beam on and off. Light beams usually just pass through each other, especially if they are relatively weak.

Researchers from the University of Toronto in Canada have figured out a way to allow beams of individual photons to affect each other, and have made a device that switches light in a manner similar to the way electrical transistors switch electrical current.

Photon transistors could pave the way for fast, low-power, all-optical computers. Extremely low-power switches are also a necessary component of quantum computers, which use the delicate differences in the states of atoms and subatomic particles to compute.

The researchers demonstrated the photon switch by shooting two weak beams of light into a crystal that was simultaneously bombarded by intense laser light of another wavelength. "The switch allows two beams of light so weak that they contain at most a single photon, and most often none at all, to meet up inside a thin optical crystal," said Aephraim Steinberg, an associate professor of physics at the University of Toronto in Canada.

One of the weird quantum traits of light is that it is simultaneously a continuous wave and a stream of tiny particles, or photons. Different colors of light are different wavelengths. Red light, for example, is around 650 nanometers, or millionths of a millimeter, from crest to trough, while higher-frequency blue light measures around 450 nanometers.

Lit up by an intense laser beam of blue light that measures half the wavelength of the weak red beams, the researchers' crystal allows weak beams of red light to pass through unless they both contain a photon. "The crystal is transparent to the two weak signal beams except when both beams contain a photon, in which case the two photons annihilate [each other], and are prevented from passing. This is the switch effect," said Steinberg.

The red color of the weak beams disappears, turning the switch off, when each contains a photon because the two photons essentially merge into one higher-energy photon of blue light, a process known as upconversion, according to Steinberg. "A single red photon doesn't possess enough energy to "turn blue" and will therefore be transmitted undisturbed," he said. "But since any pair of red photons will upconvert, it's as though a single photon is enough to switch off the path for the other photon."

The switching interaction occurs in a region of the crystal that is about one tenth of a millimeter across, but the equipment required for the researchers' prototype includes an inch-long crystal and a six-foot-wide table containing lasers and detectors. Because the actual switching is purely optical, it could in theory be miniaturized using techniques that exist today, said Steinberg.

The researchers' prototype works about 60 percent of the time, but the concept could lead to a reliable switch, according to Steinberg.

The researchers' eventual aim is to use the switch in quantum computers, Steinberg said. "Our hope is that this could be used as a fundamental logic gate inside quantum computers, whose [potential] uses are still... being discovered," said Steinberg.

Quantum computers could be much faster than the fastest possible electronic computers, because they have the potential to examine every possible answer to a problem at once. "If you know how to ask the computer the right question, instead of getting the results of just a single calculation, you may find out something about the results of all possible calculations, something the classical computer would've had to run exponentially many times to determine," Steinberg said.

The research is impressive, and "potentially very significant," said Robert Boyd, a professor of optics at the University of Rochester. "It's been well-established that a strong beam of light can be used to control another beam of light. The novel feature of the present approach is that the two weak beams interact in the presence of a strong beam, which allows the interaction to be strong even though the control and signal beams are both weak," he said.

This method has the potential to produce energy-efficient optical switches that operate with very weak power levels, which would be useful for applications like telecommunications and optical computing devices, said Boyd.

The switches are potentially useful for quantum computing for similar reasons. "The signal levels must necessarily be very weak" for quantum applications, he said.

Although there are many research efforts under way to bring quantum computing to reality, it is hard to know if and when these fantastically fast computers will materialize, said Steinberg. "Thousands of people around the world are working towards the construction of quantum computers and algorithms for use on them, but none of us knows if a full-scale device will ever work," he said. "I'd say it's equally likely that we will never see a quantum computer in our lifetimes, or that people will stumble across the right architecture for one in the next ten years or so."

Steinberg's research colleagues were Kevin J. Resch and Jeff S. Lundeen. They published the research in the November 15, 2001 issue of Physical Review Letters. The research was funded by the Canadian Natural Sciences and Engineering Research Council, Photonics Research Ontario, the Canada Fund for Innovation, the Ontario Research and Development Challenge Fund, and the U.S. Air Force.

Timeline:   > 10 years
Funding:   Government
TRN Categories:  Optical Computing, Optoelectronics and Photonics; Quantum Computing
Story Type:   News
Related Elements:  Technical paper, "Nonlinear Optics with Less Than One Photon," Physical Review Letters, September 17, 2001.


July 24/31, 2002

Page One

Disks set to go ballistic

Two-step queries bridge search and speech

Implant links nerve cells to electronics

Silicon chips set to go atomic

Light switch promises powerful computers


Research News Roundup
Research Watch blog

View from the High Ground Q&A
How It Works

RSS Feeds:
News  | Blog  | Books 

Ad links:
Buy an ad link


Ad links: Clear History

Buy an ad link

Home     Archive     Resources    Feeds     Offline Publications     Glossary
TRN Finder     Research Dir.    Events Dir.      Researchers     Bookshelf
   Contribute      Under Development     T-shirts etc.     Classifieds
Forum    Comments    Feedback     About TRN

© Copyright Technology Research News, LLC 2000-2006. All rights reserved.