Aurich Lawson / Slayer (RIP Jeff Hanneman)

These days, when physicists talk about light, they like to divide it into two categories: classical and non-classical. Of course, classical light is the boring, everyday stuff that anyone gets delivered to their doorstep roughly 12 hours every day. But non-classical light is harder to get hold of and, for physicists, obtaining non-classical light states seems to be just one step short of world domination (provided your definition of world domination involves doing quantum cryptography and quantum computing).

But, like cheap knock-off goods, genuine non-classical light can be hard to distinguish from ordinary, old fashioned classical light. Until now that is. A group of researchers, mainly from Oxford, have figured out a new way to distinguish the two brands of light. It is very clever and relatively simple. So simple that I just have to tell you all about it.

Classical/non-classical? And who cares?

The difference between classical and non-classical light states is one of statistics. Classical light sources behave one way, and non-classical light sources don't. The primary example of this is bunched and anti-bunched light.

(No, not the actual plane used.)

Gianluca Micheletti

Here in the Ars science section, we cover a lot of interesting research that may eventually lead to the sort of technology discussed in other areas of the site. In many cases, that sort of deployment will be years away (assuming it ever happens). But in a couple of fields, the rapid pace of proof-of-principle demonstrations hints that commercialization isn't too far beyond the horizon.

One of these areas is quantum key distribution between places that aren't in close proximity. Quantum keys hold the promise of creating a unique, disposable key on demand in such a way that any attempts to eavesdrop will quickly become obvious. We know how to do this over relatively short distances using fiber optic cables, so the basic technique is well-established. Throughout the past couple of years, researchers have been getting rid of the cables: first by sending quantum information across a lake, then by exchanging it between two islands.

The latter feat involved a distance of 144km, which is getting closer to the sorts of altitudes occupied by satellites. But exchanging keys with satellites would seem to add a significant challenge—they move. Over the weekend, Nature Photonics published a paper that indicates we shouldn't necessarily view that as an obstacle. The paper describes a team of German researchers who managed to obtain quantum keys transmitted from a moving aircraft.

Yang Juan

A new system called the EmDrive, a way of using electricity to generate thrust without the need for fuel, is one of those ideas that will generate a lot of heat and noise, but probably not a lot of thrust for many years to come. I had never heard of the idea until today, and the latest paper, a translation, doesn't throw a lot of light on the physics itself. So brace yourself and let's see what we can figure out from a grainy line drawing in the translation.

The idea is based on a standing wave cavity: two mirrors—in this case, microwave mirrors—facing each other so that microwaves travel back and forth between them. In one picture, two waves are travelling in opposite directions between two mirrors. If you look at both waves simultaneously, however, you get another picture: a standing wave, like the vibration on a violin string. The key thing about this system is that these two pictures must be self-consistent. We will use that to examine the EmDrive's potential for thrust.

In our simple picture of two mirrors facing each other, the mirrors are subject to equal and opposite forces. Essentially, the photons reflect from each mirror, exerting a force in doing so. Because this is symmetric, the cavity cannot have a net force. Even if we were to make the light very focused at one end and very diffuse at the other, we are simply distributing the same number of photons differently at each end, so the total force remains the same.

This metamaterial is the aperture of the new microwave imaging device.

John Hunt

Add image compression to the list of nifty applications for metamaterials. Metamaterials guide light waves to create “invisibility cloaks” and bend sound waves to make theoretical noise reduction systems for urban areas. But these materials are tuned to particular wavelengths; some invisibility cloaks don’t work at all visible wavelengths because they leak those wavelengths of light. Now researchers have capitalized on that leakiness to build a new functional device: a microwave imaging system that compresses an image as it's being collected—not afterward as our digital cameras do.

Every pixel in a picture from our digital cameras corresponds to a pixel of information recorded on the detector inside the camera. Once a camera collects all the light intensity information from a scene, it promptly discards some of it and compresses the file into a jpeg (unless you explicitly tell it to save raw data). You still end up with a decent picture, though, because most of the discarded data was redundant.

Compressive sensing aims to ease this process by reducing the amount of data collected in the first place. One way to do this is with a single pixel camera, developed in 2006. These devices capture information from random patterns of pixels around the image, essentially adding the light intensity values of several pixels together. If you know something about the structure of that image—say clusters of bright stars set against a dark sky—you’ll be able to capture that image with fewer measurements than a traditional camera.

Electron micrograph of a nanophotonic phased array (NPA), consisting of 4,096 antennas. The image on the right zooms in on one of these antennas, which consists of a small set of lithographed ridges and a small silicon waveguide.

Jie Sun

A single antenna—like a single ear, eye, or audio speaker—works, but combining two or more are better. This works for sending as well as receiving. By sending the same signal from two or more identical antennas in phase with each other, you can boost it, make it extremely directional, or change its shape. This is the concept behind phased arrays, which have found great success in communications, radar, and radio astronomy.

Until now, phased arrays have mostly utilized radio frequencies. Modern nanoscale technology is now allowing researchers to create phased arrays for optical (visible) light. Jie Sun and colleagues fabricated a phased array of 4,096 microscopic antennas on a single silicon chip. This allowed them to shape the output waveform, so they could transmit an image of the MIT logo by combining the light from each tiny antenna in precise ways—something that could not be done with (say) a similar array of LEDs. Potential applications for this research include biomedical imaging, holography, and laser communications.

Phased arrays have had their greatest success at radio wavelengths, where they have been used for radar, AM radio broadcasting, and so forth. (In radio astronomy, the technique goes under the name "aperture synthesis" and led in part to a Nobel Prize for Martin Ryle and Anthony Hewish.) The idea in all these cases is the same: begin with one signal, but multiple identical antennas. By combining the output from each antenna, researchers can selectively create interference patterns, boosting the signal at distance or canceling it out entirely where it might overlap with another signal.

One of the two devices used to demonstrate that boson sampling works in the lab.

Jame C. Gates

We've covered a lot of the progress that has gone into creating effective quantum computers. Although there has been a good deal of progress in terms of building individual components, we haven't been able to put things together into a complete device. That has left us building small, proof-of-concept systems that are handily outperformed by existing classical systems. Without a large enough system, we can't clearly demonstrate the sort of accelerated performance we predict we'll get from quantum hardware.

Two groups of researchers have now figured out a way to test whether quantum systems really can outperform a classical computer. Their systems, which take advantage of a phenomenon called boson sampling, can't be used to compute any algorithms, so they're not as useful as quantum computers might be. But they can be used to confirm that we're on the right track when it comes to quantum computers.

A quantum computer isn't like our existing computers, where electrons flow through a series of switches. Instead, a carefully prepared quantum system is allowed to evolve, and it is then measured. The system only provides us with an answer because we can map different answers to all the possible states that the system can end up in. Because quantum systems evolve very quickly, it should be possible for these systems to arrive at an answer much faster than a typical computer.

Warner Bros. / Aurich Lawson

Take a very thin sheet of metal and drill it with tiny holes in a regular rectangular pattern. Ordinarily, if you shine light with wavelength that's larger than the holes, it wouldn't get through—the metal would be opaque. However, in the case of this particular pattern of holes, a lot of the light gets through the sheet anyway, a phenomenon known as extraordinary optical tranmission (EOT). Since the discovery of EOT, the effect has been harnessed in a number of optical and biophysical devices. A full theoretical understanding of the phenomenon proved elusive, however, which could hamper further device development.

A systematic exploration of hole spacing may help elucidate the mechanism behind EOT. Frerik van Beijnum and colleagues demonstrated that electrons on the metal's surface have two properties that contribute to EOT, with different strengths depending on the hole density and configuration. These results enabled the researchers to determine the physical parameters that dictate EOT, potentially leading to new device designs.

Ordinarily, light can pass through an opaque barrier only if the barrier is pierced with openings larger than the light's wavelength. (This also applies to all manner of waves, including sound and water waves.) That's why EOT is fascinating: the holes are smaller than the wavelength, yet a substantial amount of light still gets through something that would ordinarily be opaque. Oddly, making the material thinner—and therefore more transparent—decreases the EOT effect.