THE BEST WAY to communicate in secret is to do it in the most public place possible. If your message is in code and the code is unbreakable, you can shout it from your window, you can put it on a billboard, you can even broadcast it on Oprah if you like. Spies around the world can watch the show, but if they don't have the key to your code they'll understand nothing, just like everyone else. The problem, however, is that at some point you have to give the key to your code to the intended recipient of your message. You can't put the key into code. That's just deferring the problem because you'll have to give the recipient the key for that as well. You can't do it in public because someone might be listening. If a spy gets the key, they crack the code. If they crack the code, they get the message, and if they get the message your secret is out.

Quantum cryptography is the only way to ensure the absolute safety of a key. This is because the key will be protected by the laws of physics. The implementation of quantum communication, of which quantum cryptography is one type, is in its infancy. Until now, it has been experimentally demonstrated only over very small distances (tens of kilometers). Yet in today's issue of Nature, Professor Anton Zeilinger and colleagues at the Institute for Experimental Physics in Vienna propose a method that could increase the distance of quantum communication tenfold.

The idea of interacting via the quantum realm is as appealing as it is strange. Essentially, it involves communication by pure energy. Photons -- light particles -- can take on one of four polarizations or states. Such relatively simple changes of state can be exploited to convey one of four pieces of information (e.g., one state signifies "1," another "0," etc.). If you string a bunch of photons together, a relatively complicated message, like the key to a code, can be conveyed.

What makes quantum communication perfect for cryptography is Heisenberg's Uncertainty Principle. The principle states that the observation of a quantum system necessarily changes that system. It is an extremely odd, but nevertheless fundamental property of photons that when unobserved they exist in a state of undecidedness. When they are observed, however, the very act of observation forces them to assume a well-defined state, which in this case could be a specific polarization. This means that the spied-upon photons will carry an indelible record of any third-party observation. By comparing a small sample of the received information with the sender, the intended recipient of the key can determine if the state of the photons has been significantly altered. They may also find that the structure of the message has been changed, because it's not possible for an eavesdropper to intercept a quantum transmission channel without absorbing some of the photons.

In fact, the mark left by an eavesdropper is proportional to the amount they have eavesdropped. So, even though quantum cryptography can't ensure that someone won't try to spy on you, it does mean that the recipient of your key can always tell how much a third party has been listening in. And once you know someone has been eavesdropping enough to determine what the key is, you can pick another key and try again.

How to isolate photons for a message from the abundance of light is one of the problems of quantum information systems. "A rough estimate of the amount of light particles in a well lit room at any point," says Zeilinger, "is 10¹⁵ [one thousand million million], that's how small these buggers are." But for quantum cryptography you need just a handful. How to transmit that handful once you have them is another challenge. Photons can be transmitted through air using lasers or via fiber-optic channels. But they are extremely fickle. This means that their state may change over distance or they may be absorbed in the transmission channel. While a vast majority of the photons pass through air or the glass of a fiber-optic tube successfully, minor impurities contained in the glass or air can absorb them. The longer the distance between sender and receiver, the greater the number of impurities, and the more likely it is that the photons will be lost or altered. 

There are a few types of quantum communication. The type that Zeilinger and colleagues propose to use exploits the entanglement of photons, which Zeilinger calls "the most curious quantum phenomenon there is." Entanglement means that two (or more) photons mimic each other's state. For example, if one becomes polarized in a particular direction, the other will be, too. What's curious about this relationship is that it's non-local. Photons in an entangled pair could theoretically be miles apart, but if you treat one, the other reacts correspondingly. So you might have one photon of an entangled pair in New York and the person you want to communicate with has the other in L.A., and by effecting a change in the New York photon you simultaneously effect a change in the L.A. photon. In this way, it's possible to communicate at that instant without transmitting a thing.

There are various ways to produce entangled photons, and at this stage, says Zeilinger, "it's a fairly random process." Because entanglement degrades over time, it's necessary for long-distance communication that a pair be as perfectly entangled as possible to begin with. Zeilinger's proposed mechanism, a polarized beam splitter (PBS), is a little cube of gas that generates highly entangled states in photons that are only weakly entangled, making them robust for long-distance communication. While there are no specific plans to build the PBS yet, Zeilinger's team is currently bidding on a number of industrial models of quantum communication devices being built in Europe; if they win the bid, the models will presumably incorporate some kind of PBS device. The PBS would replace an earlier proposed device known as a CNOT gate. It would achieve only 50 percent of the predicted results of the CNOT gate, but the PBS is still a distinct improvement: According to Zeilinger, it's "well within the reach of current technology." As is the way with theoretical physics, the CNOT gate is a device that doesn't exist, and no one knows how to build it.
Christine Kenneally is an Australian writer who lives in New York.
Other articles by Christine Kenneally