A TRICK of the light has allowed us to bend the rules of quantum mechanics. This may one day prove useful for building quantum computers.
Many of our intuitions about quantum mechanics are based on a foundation of experiments using just one or two particles of light, called photons. One of the most famous is the double-slit experiment, in which a single photon is sent towards a barrier with two slits in it. Classical physics says the photon can only pass through one slit, but quantum physics says otherwise: the photon creates an interference pattern as if it has gone through both slits.
This can only happen if you don’t attempt to measure which slit the photon passes through: trying to have a peek by placing a detector at the barrier prevents the pattern from forming.
Advertisement
“There’s this magic-seeming thing that happens in quantum mechanics, which is that if there are multiple ways for an event to happen and you don’t know which way it happened, you get quantum interference,” says Alex Jones at the University of Bristol, UK.
Now Jones and his colleagues have discovered that adding more photons can bend the no-peeking rule. Another key experiment uses two photons that can each travel along two different optical fibre paths, a total of four possible outcomes. When the photons are indistinguishable from one another, quantum interference means they bunch together and always take the same path.
If they are distinguishable, though – for example, if one is a red wavelength and the other is blue – they sometimes take different paths. The more distinct they are – the further apart their wavelengths, say – the more likely this is to happen.
Jones and his colleagues performed a variant of this experiment with four photons, each with four possible paths, a total of 16 ways. Instead of making the photons red or blue, the team manipulated their polarisations and the times at which the photons were sent into the experiment.
The group found that even though the photons were different, they all interfered with each other, meaning they followed the same path more often than we would expect based on classical physics. “For large-enough systems, the intuition you get from small-scale demonstrations with just two photons breaks down,” says Jones. It should work with more than four particles, he says.
This doesn’t break any rules of physics, says Jones. It works because there is still a degree of uncertainty over which path each photon took, which causes quantum interference in the same way as not being able to tell which photon is which ().
Barry Sanders at the University of Calgary in Canada is sceptical. “I don’t think this is a surprise,” he says. “The way we typically talk about photons interfering, we don’t typically take into account polarisation.” That could be introducing a spurious result, he says.
If the experiment holds up under scrutiny, Jones says that the practical uses aren’t clear yet. “The original two-photon experiment is one of the most important parts of the toolkit for some kinds of quantum computing,” he says. “Maybe in some years’ time our work will find a practical application in these quantum technologies too.”
