BY DAWN LEVY
Information flooding the ever-evolving Internet will increasingly take the form of light pulses streaming through fiber-optic cables. But fiber optics brings both solutions and problems. Conceivably, hackers could use beam splitters to divert streams of light and access confidential information without being detected. But if a message were carried by a lone photon -- the smallest discrete quantity of light, called a quantum -- it would be easier to detect intruders.
"If you have only one photon per pulse, you would immediately know that an eavesdropper had penetrated the system because the receiver at the opposite end could tell that the data had been disturbed," says chemistry Professor W. E. Moerner.
He and visiting research associate Dr. Brahim Lounis, now at the Universite Bordeaux in France, were the first to use lasers to get single molecules to emit single photons on demand at room temperature. The achievement, published in the Sept. 28 issue of the journal Nature, takes cyberspace a quantum leap closer to secure communications.
While quantum communication is still futuristic, Moerner says, it aims to provide the ultimate in secure information transfer. In the next five or 10 years we may use quantum information technology to send messages over channels one photon at a time. Or we may employ quantum cryptography, which uses signals from a single photon to transmit an electronic "key" to decode encrypted messages.
"You want to minimize the probability of emitting two [or more] photons for every pulse because that would allow an eavesdropper to split off one of those photons and read your key without your knowing it," Moerner says.
Until Moerner and Lounis's work, the only way to coax single photons from a pulsed laser at room temperature was by attenuation -- that is, making the laser beam weaker and weaker until each pulse carried only a small number of photons.
But attenuation is an inefficient way to produce single photons. "You'll mostly get zero photons per pulse and then a small probability of one photon per pulse and an even smaller probability of two photons per pulse," Moerner says.
Moerner and Lounis's system is much more efficient. It can produce single photons 86 percent of the time. It produces no photons 14 percent of the time, and two photons, hardly ever. That nearly zero probability of two-photon emission could be used to ensure the immunity of quantum communications against hacker attacks.
But is getting a single photon 86 percent of the time good enough for communications applications? "It's far, far better than what [researchers] have now," Moerner says. "It's a complex issue of what 'good enough' for applications is, but certainly big improvements in efficiency and simplicity help. What they typically work with now is single-photon probabilities around 10 percent or less."
The system in essence turns a single molecule into a light source. But it's not a classical light source like a lamp, where the light-bulb filament heats up and many photons boil chaotically off the filament surface. This light source is quantum-mechanical, emitting only one photon at a time.
"The beauty of this whole idea is that it's so straightforward," Moerner says. "A simple room-temperature apparatus can generate this quantum mechanical light source."
How does it work? Short, fast pulses of infrared light shoot out of a laser. But red is not an energetic enough color (wavelength) for the purposes of the experiment, so the light next travels through a device called a second harmonic generator that halves its wavelength and doubles its energy. The light that emerges is green, which is energetic enough for the experiment. The light then enters a scanning confocal microscope, which focuses the beam on a thin crystal flake. The flake is made of a very small number of single terrylene molecules embedded in a crystalline slab of p-terphenyl molecules. By moving the laser, scientists can aim the beam at a single terrylene molecule. Light hits the single molecule with the right amount of energy to "pump" it from its ground state to its excited state, causing it to -- voilà! -- release a single photon.
Moerner, previously at the University of California-San Diego for three and a half years, and IBM Almaden for 13 years before that, has been at Stanford only a year and a half. He funded the project partly with Stanford money and used equipment that was purchased with a prior grant from the National Science Foundation.
"We like to explore single molecules any way we can," says Moerner, whose group of six graduate students and postdoctoral scholars currently is concentrating on studying the behavior of single protein molecules.
In 1989, his group at IBM was the first to use lasers to select, probe and measure the properties of individual molecules. Similarly, Steve Chu, a professor in the departments of Physics and Applied Physics, and his research group have used lasers to study the characteristics of single polymers, or chains of repeating molecules. Other Stanford researchers using lasers to study single molecules include Professors Richard Zare (chemistry), Steven Block (applied physics and biological sciences) and James Spudich (biochemistry and developmental biology).
And one other Stanford researcher uses lasers to produce lone photons. But his method doesn't use single molecules. At extremely low temperatures (near absolute zero), Yoshihisa Yamamoto, professor of electrical engineering and applied physics, has created an electronic "turnstile" that allows passage of a single photon -- or multiple photons if desired -- in a controllable way.
sources may be implemented in real-world communications systems in
five to 10 years, Moerner predicts: "There are a number of research
projects at major labs. There's been a group at IBM, for example,
working on a quantum communication channel. There's no product at
the moment, but it's in the development stage."
Stanford Report, November 8, 2000