Giveaway clicks

[Bartley et al. Direct Observation of Sub-Binomial Light, Phys. Rev. Lett. 110 173602 (2013)]


At the beginning of the 20th century, physicists understood light as an electromagnetic wave, which helped to understand optics and electricity and magnetism using a single picture. But the equations also predicted that every object would radiate an infinite amount of energy, unless they were frozen at absolute zero! Eventually this huge problem was fixed by Max Planck, who guessed that light could only be emitted in lumps, which today are called photons. This was the beginning of quantum mechanics, but even though the story began with light, the quantum theory only applied to the way light was emitted, which is to do with atoms. So is light really made of indivisible particles called photons, or is it just that atoms only spit out waves in bite-size pieces? Do we really need a quantum theory of light, or just a quantum theory of matter?

Perhaps not surprisingly, we now understand that in fact light is a quantum object, and photons are intrinsic to its structure. But how do we see this? When we shine light on an ultra-sensitive detector, we get a series of “clicks”, so it seems like we can measure each photon hitting the detector (even the human eye can detect such weak pulses of light, although our brains filter them out as noise). But in fact these clicks are what you would expect if the atoms in the detector were quantum, so this isn’t really evidence for photons.

Actually, counting clicks CAN tell you about photons, but you have to look carefully at the statistics of the clicks. Not just how often they come, but also how often you get two clicks versus one click. Or three versus two. It was recently shown [Sperling, Vogel & Agarwal, Sub-binomial Light, Phys. Rev. Lett. 109 093601 (2012)] that when using an array of many detectors, the clicks would always follow a binomial distribution if the light is not quantum (known as classical light). If you ever see different statistics -- if you see sub-binomial light -- then this means you are looking at non-classical light, which can only be understood by thinking about photons. Sub-binomial statistics are therefore the signature of the quantum nature of light.

Now researchers at the University of Oxford [Bartley et al. Direct Observation of Sub-Binomial Light, Phys. Rev. Lett. 110 173602 (2013)] have implemented this test for quantumness by splitting a light pulse into many smaller pieces to replicate the effect of using an array of detectors. As expected, when they used classical light from an ordinary laser, they saw binomial click statistics, but when they instead used the light emitted by a non-linear crystal -- which is well-known to be non-classical -- they saw the statistics change dramatically to a sub-binomial distribution, with a much smaller variation in the click rate than they saw with the classical light.

This experiment is certainly not the first observation of photons or non-classical light. These concepts are the bread and butter of modern quantum optics research. But the binomiality test is very convenient because no special knowledge about the detectors being used is required to apply the test, which makes is much easier to certify when light is behaving in a quantum way. The Oxford group hopes to apply their test on larger networks of quantum light states, which could be useful as a benchmarking tool for future quantum technologies based on encoding quantum information in light.

This work was undertaken as part of the Q-ESSENCE project.

Project & realization: Pixels United.
Copyright QUIE2T 2010 - 2012. All rights reserved.