Two articles for Wired.com on early developments (circa 2001-'02) in quantum cryptography
1) "Quantum Crypto to the Rescue," Sept. 7, 2001
2) "A New Twist on Light Speed," June 26, 2002
3) PLUS this feature story from the April 15, 2006 issue of New Scientist magazine:
Missing in action
When photons go AWOL, it should be curtains for your quantum computer. So how come it works better without them, asks Mark Anderson
• JAMES Franson is no stranger to strange things. He has done plenty of experiments showing, for example, that quantum entanglement really is as spooky as it seems. Einstein hated the idea, but Franson has seen at first hand that two quantum particles, such as photons, can affect each other's properties no matter how far apart they might be.
Even that degree of experience in negotiating quantum weirdness, however, did not prepare Franson for what he has just found. It turns out that it is not just quantum particles that get entangled: if Franson's calculations are correct, you can entangle bits of empty space as well. Einstein would really have been spooked by this one.
This extraordinary discovery arises from Franson's efforts to make particles of light, photons, more reliable carriers of information. Researchers in the field of quantum information use the quantum states of individual photons to encode binary digits, or bits: changing the angle of a photon's spin axis, for example, can make it encode a 0 or a 1. The trouble is, to make this trick useful, you must be able to send those photons around your communications network or your quantum computer — and that's far from straightforward.
Photons have a nasty habit of disappearing as they travel through an optical fibre. Or they might hit an impurity on their travels and have their spin knocked out of alignment. That's a big problem for quantum communications because quantum systems rely on each photon carrying its own bit: lose that photon, and you may lose a vital part of your calculation or communication. Classical communication using light is less prone to that problem because it encodes the information on a pulse containing lots of photons.
Franson made his discovery while he was working with a "swap gate", a building block for quantum information processing. A swap gate is a kind of crossover junction. Two fibres move close together and can swap their signals, so photons initially travelling on one fibre move into the other . Franson noticed that each time the gate failed it was always for the same reason: one photon would cross over to the other fibre while the other one continued on its original path. That meant he had two photons travelling through the same fibre. "We realised that all the failure events corresponded to situations where two photons would come out of one fibre," he says.
To overcome this problem, Franson decided to employ something called the quantum Zeno effect. "The Zeno effect is somewhat analogous to the notion that a watched pot never boils," Franson says. "This is actually true in quantum mechanics."
Franson's plan was to incorporate a "watcher" into the optical fibres. "If we could watch for the presence of two photons in an optical fibre often enough, the failures would be eliminated," he says.
That may sound strange, and it is — very. The watcher works because a photon's quantum state cannot change or evolve if the photon is observed frequently enough. "In the quantum Zeno effect, a randomly occurring event can be suppressed by frequent observations to determine whether or not it has occurred," Franson says.Watch duty
The randomly occurring event, in this case, is the move of just one photon from one fibre to the other, a move that changes the quantum state of the photon. The watcher in this scheme comes in the shape of a carefully chosen atom which, because of the arrangement of its energy levels, absorbs two photons far more easily than it ever absorbs just one. "Dope" the glass that makes up the optical fibres with a series of these atoms, and continuous "observation" is effectively taking place as the photons travel along the fibres. The result is that two photons will never co-exist in the same fibre.
The twist came when Franson wondered what would happen if, instead of using single photons, he fired a laser beam into each of the fibres. Now there would be a stream of photons moving in lockstep through the fibres, rather like a column of soldiers on parade. When they passed the two-photon absorbing atoms, however, some absorption would be inevitable. The beam of photons would continue to stream through the fibre, except now it would contain "holes", as if a few soldiers had been plucked at random from the parade.
Once this has happened, the photon stream has three rather strange attributes, Franson says. First, these holes behave like particles in their own right. This is a well-known concept in semiconductor electronics, in which the holes left by electrons "missing" from their usual position within a material can be treated as positively charged carriers of current. That's analogous to what happens with photon holes, Franson says.
The second discovery is more surprising: the holes will be in a "superposition" — they do not exist in a definite place within the beam, but are in several places simultaneously and will only adopt one definite position when their position is measured.
The third discovery is just plain weird: the holes in the two different beams are "entangled". That's to say, even though they are simply regions of empty space — chunks of nothing — they share a spooky quantum link.
In some ways, Franson points out, the entanglement between the holes could have been predicted because the process of their creation is the inverse of the process physicists use to create entangled photons. If you want to create a pair of entangled particles, the most common method is to shine a laser beam into a "non-linear" crystal, one whose optical properties depend on its orientation. If the alignment of the crystal is just right, each photon in the laser beam splits in half, creating a pair of entangled photons. This happens because of the way the atoms in the crystal absorb and re-emit energy. Although Franson's watcher atom absorbs two photons without re-emitting anything, it is still a quantum mechanical process, and entanglement is the outcome of all such processes — even when the products are chunks of empty space.
Nonetheless, Franson admits, it is quite a turn-up that the same is true of the photon holes. "Although some follows from what came before, a lot of this did take me by surprise," Franson says.Commanding from afar
The analogy with semiconductors set Franson wondering whether the photon hole might be just as useful as the electron hole is in computer technology today — especially when you add entanglement into the mix.
The beauty of quantum entanglement is that you don't necessarily even have to encode any information on the quantum bits being transmitted. If you have an entangled pair of qubits, you can send each of them off to their respective destinations, then do something to one that effectively puts the other one into a particular quantum state. "These kinds of correlations can be used to implement secure communications — quantum cryptography — or to transmit an unknown state via quantum teleportation," Franson says.
Though it is normally the preserve of photons, the holes could be used as qubits in exactly the same way. To make the best use of entanglement, you would want the holes to be far apart, so rather than firing two beams in the same direction, Franson envisages a scheme in which the laser beams travel through the fibres in opposite directions. In this setup, the entangled holes will soon become widely separated. When you measure where hole number one actually lies along the length of the laser pulse, its superposition will collapse into a definite position, along with that of hole number two within its own laser pulse. Though he doesn't yet have a scheme in mind, Franson thinks this discovery could lead to a new way of processing quantum information.
Partly, that is simply because the holes, which are preserved by the lockstep arrangement of photons in the laser beam, might well be more robust than single photons. The absence of a photon will be far less affected by collisions with impurities in the fibre-optic cable than any individual photon propagating through the fibre would be. "It may be that the photon holes are less susceptible to such effects," Franson says. And if they are more robust than photons, their entanglement could be put to better use.
What can really be done with the entangled holes may depend on their precise nature. Jonathan Dowling of Louisiana State University suggests the entanglement being carried forward by the photon hole is not quite what it seems, but is actually a result of small entanglements between all the unabsorbed photons surrounding the hole. Franson is not sure about this interpretation, however. Those surrounding photons are all emitted at different times by the lasers, he points out, and are unaffected by the photon absorption. "It is difficult to understand how the surrounding photons could become entangled," he says. "It is easier to think of the holes as being created in an entangled state."
There is still a way to go before we know the exact nature of the entanglement. Although the idea has been published in Physical Review Letters (vol 96, p 090402), no photon hole experiments have been done. According to Paul Kwiat of the University of Illinois, the difference between theory and experiment in this case may be very large indeed. "I think in principle it can be built," Kwiat said. "In practice… maybe."
Kwiat adds that, even if you could devise the experiment, to produce entangled photon holes and then have them travel apart from each other would require the two incident photons, travelling in opposite directions, to arrive at the two-photon-absorbing atoms at precisely the same instant. And then there's the problem of detecting the holes. Holes themselves can't be collected by a photon detector; only an unusual absence of photons could be observed. "When you actually look at the rate of detections on a photon counter that's clicking away, they're coming at random times anyway," Kwiat says. "So since they're coming in at random, noticing that one of them is missing can be difficult."
Franson knows it will be challenging, but it's worth pursuing: when it comes to quantum computing, there are times when nothing is definitely better than something.
"The absence of a photon will be far less affected by impurities in the fibre than any photon"QUANTUM TRICKS WITH EMPTY SPACE How to entangle two chunks of nothing
Photons travelling through two close and parallel fibres will eventually tunnel across into each other's fibres. This "quantum swap gate" can be used to build logic gates tor quantum communication
When a swap gate goes wrong, two photons end up in the same fibre. This can be avoided by doping each fibre with atoms that will absorb photons in pairs, but not singly. Each atom effectively "measures" the quantum states of the photons as they pass. This pushes each photon back into its initial state within its original fibre
Fire two laser beams towards each other in this doped fibre, and some two-photon absorption will occur, leaving holes in the beams. Each hole exists in a "superposition" of places within its beam1 and is entangled with the hole in the other beam. Perform a position measurement on one hole, and the other is forced into a specific location within its beam Entangled holes could provide a useful new tool for quantum information researchers
By Mark Anderson
(cc) image by Giorgio Brida