January 19, 2001 3:35 PM PST
Stopping light could lead to quantum advance in computing
- Related Stories
For 2001, futurists are being a bit on the shy sideDecember 28, 2000
The experiments were conducted by two teams working independently of each other in Cambridge, Mass. One team was led by Lene Hau of Harvard University and the Rowland Institute of Science, the other by Ronald Walsworth and Mikhail Lukin of the Harvard-Smithsonian Center for Astrophysics.
Just a few years ago, Hau brought the speed of light down to a manageable 30 meters per second, much slower than its normal 300 million meters per second, according to Seth Lloyd, an associate professor at the Massachusetts Institute of Technology whose focus is on building quantum computers.
More recently, the two teams brought light to a halt by shining a pulse of light into a chamber of gas in which the beam got slower and slower and dimmer and dimmer before coming to a stop with the help of a technique dubbed electromagnetically induced transparency. Hua's group used chilled sodium gas to act as a parachute, while Walsworth's team used gaseous rubidium, an alkaline metal element.
The light, composed of particles called photons, essentially lost its zing as the information from the photons was transferred into the spin inherent in the gas atoms. Once paused, it could then be revived to its usual speed of 186,000 miles per second.
The achievement has sparked renewed enthusiasm among advocates of quantum computing.
"It is easy to send a photon from one place to another, but catching it at the other end is what is really hard," Lloyd said. "This is a beautiful way of catching bits stored on light and storing them in a medium. I think it puts us considerably forward in our schedule in building more powerful quantum computers and the quantum Internet."
Traditional drives in today's computers store data in terms of zeros and ones, with different combinations and strings of zeros and ones representing information.
"The problem with classical computers is that they can be zero or one only at a given moment," said Ben Stein, a senior science writer at the American Institute of Physics. "Quantum computers would use particles that act as a zero or one at the same time, giving them the ability to perform many, many calculations in parallel, while classic computers could only look at one possibility."
Physicists envision a quantum computer could perform calculations that are exponentially faster than any computer now available, cracking the toughest encryption code in a matter of seconds and searching massive databases in a fraction of the time it takes today.
A top-notch quantum computer can process data in 7- or 8-bit chunks and perform a few thousand operations.
"That sounds pretty pathetic, but just five years ago we had a 2-bit computer on which we could do just one operation, and seven years ago, it had no bits at all," Lloyd said. "It would be irresponsible to speculate when we could have even more powerful quantum computers based on these developments, but this is really going to help."
Lloyd said that one of the hardest things in quantum computing is the transfer of information from light to atoms. Making light stand still so it can be looked at and manipulated will "be quite useful in building quantum computers."
Both MIT and Caltech are trying to build communication applications based on quantum physics, such as a quantum Internet that would have little quantum computers attached to each other with fiber optics sending and receiving information using photons.
Hau's experiment will be published in the journal Nature, and Walsworth's work will appear in the Physical Review Letters.