Unbreakable codes. Computers powered by photons. Until recently, this was the stuff of spy movies and science fiction. But not anymore.
Three faculty members in UCSB’s Electrical and Computer Engineering Dept., Atac Imamoglu, Evelyn Hu and Pierre Petroff, reported building a device which emits individual photons – indivisible packets of light – on demand. This machine, developed over two years and announced in December, could be the first step toward building a new type of computer that operates using photons instead of electrical current. Photon computers would be many times faster than the current electric versions, although Petroff said the application is still five years away.
Single photon emission also makes possible a method of encoding secret messages, known as quantum cryptography. This system is based on the idea that photons are the most basic unit of light and cannot be divided. A system of codes based on these basic units could contribute greatly to national security as well as domestic privacy.
The basic idea, Hu said, is that photons carry information. “The question about encryption is when you send information, how do you send it so that only the person for whom the message is intended will read it the right way?” she said. “You do this by sending a code within that message.”
It is impossible to observe a photon individually, because to see something requires bouncing a photon off of it, and bouncing a photon off another photon makes them both change course. The act of observing a photon changes the message, Hu said.
“It’s like steaming open an envelope, looking at the letter inside and then closing it so no one knows what’s going on,” she said. “[If] you only have a single photon that’s carrying that information and somebody wants to know what’s in the message or what’s in the key, they have to access that photon. But as soon as they access that photon, then we know that it’s been tampered with. And we know that security has been violated.”
The Federal Office of Naval Research funded the bulk of the research, but it may have benefits that reach beyond national security and into the domestic realm.
“This makes the FBI’s wish to be able to decrypt any message sent by any individual or company that much harder,” said Lawrence Badash, a UCSB history of science professor and board member of the American Civil Liberties Union’s local chapter. “This issue had raised civil liberties questions about ‘Big Brother’ being intrusive in people’s lives.”
Beyond quantum cryptography, Immamoglu said his group plans to explore applications for the invention in quantum computation.
A normal computer uses electrons or lack of electrons to represent the ones and zeros of binary code. The idea behind a quantum computer is that a machine can be built which uses the presence or absence of light to function as ones and zeros. Digits such as a one or a zero are known as qubits. Further down the road, it has been proposed that computers will use different polarizations (wave directions) of light as qubits.
“The applications for this are probably five years from now if everything turns out well. For now, there are other experiments to try to show that we can in fact perform quantum computing and quantum cryptography,” Petroff said. “And that’s going to take maybe a year or so. So we should have some demonstration of quantum cryptography by then. Quantum computing is going to be much, much, much, much harder.”
Whatever its implications, the key to succeeding with quantum cryptography is finding a source of single photons. Hu’s research group built a circular disk called a microdisk resonator, which does the bulk of this work.
Inside this disk are distributed small optical sources called quantum dots, fabricated by a technique called Molecular Beam Epitaxy. The technique for making these quantum dots was first discovered by Petroff. Quantum dots are also known in the physics world as artificial atoms. The complete structure looks something like a mushroom.
Though the device is complicated, problems have left the professors undaunted. “We’ve been remarkably lucky in this collaboration,” Hu said. “We’ve gotten a bunch of great results already.”
ECE grad student Alper Kiraz was present when lab results showed the machine was working.
“When the proof came back for the single photon generator, we were excited in the lab, I can say. Although at the very first stage you are never sure the results will be accepted, we were feeling good about that because the results were showing up in a good manner,” Kiraz said.
All of the professors emphasized that this project was the result of a true collaborative effort, characteristic of the university. “UCSB,” Immamoglu said, “is the only place that has strength in all aspects of semiconductor nanocrystals – from growth to theory.”