Photons are easy to produce, at least en masse. But making them one at a time in a controlled fashion is much harder. Until recently the only trick physicists had for this was to reduce the brightness of a beam until it contained only one photon at a time, on average.
Of course, the “on average” clause allows for any manner of multiphoton sin: the photons may be produced in pairs or bunches or not at all. That’s a serious problem for techniques such as quantum cryptography because any extra photons are equivalent to a leak of information that an eavesdropper can use to crack the code.
Enter the photon gun brigade, a growing band of physicists attempting to build devices that can fire identical, single photons on demand. The single photon part of this problem has been cracked by a number of groups who have built guns that work by inducing atoms, ions, molecules or quantum dots to fluoresce in a cavity.
What has turned out to be more difficult, but just as important, is making each of these single photons identical. What tends to happen is that the cavites vibrate in various ways and this subtly distorts the wavelength of the photons they produce. The result is that each photon is very slightly different, which spoils their ability to interfere and become entangled with each other.
Here’s the breakthrough: Andrew Shields, at Toshiba’s research labs in Cambridge, UK, plus a few pals from a nearby university, say they can create the photons using a quantum dot that they zap with two precisely timed voltage pulses. The first pulse injects charge carriers into the diode while the second suddenly shifts the quantum dot’s emission characteristics so that it can emit light. When that happens, the dot emits a photon of a specific energy.
But the key is to use such short voltage pulses that the photon doesn’t have time to be influenced by its surroundings. And that means the quantum dot always emits photons of precisely the same wavelength.
So it looks as if Shields and co have a neat new trick up their sleeves, albeit one that has to be performed at 4K. But it’s also an entirely new mechanism for producing photons of this kind so improvements are gonna be forthcoming. How long before we see a commercial version?
Ref: arxiv.org/abs/0803.3700: Indistinguishable Photons from a Diode