Curved light drives boffin one-upmanship

For four years, scientists have been on an international race to exploit a property of a peculiar kind of light called an “Airy beam”. Now, not one, but two groups of researchers are laying claim to the prize.

The beam, named after 19th century scientist Sir George Biddell Airy, was first observed in 2007. At the time, it was predicted that a particular characteristic of the Airy beam, its ability to “self-bend during propagation while remaining invariant in shape”, would allow them to break one of the basic rules of light, that it travels in straight lines*.

The benefit of being able to bend – and particularly, control the bending – of light is in applications like the optical computer. For example, instead of having to find ways to fabricate waveguides at the small scales of computer chips, it might be possible to use “bendable light” to route some signals.

That makes Airy beams a valuable research target for researchers all over the world. Over the weekend, two groups made very similar announcements, almost at the same time: the Australian National University, and Lawrence Berkeley National Laboratory.

(The releases are dated, respectively, August 12 and August 11; given that one is in Australia and the other the US, that makes them pretty much simultaneous).

The ANU group demonstrated the ability of an Airy beam to be “bound” to the flat surface of a chip, and in that state, to travel around an obstacle on the chip and resume its original course once past the obstacle.

Professor Yuri Kivshar, a member of the ANU team, called this property of the Airy beam “self-healing”. A colleague of Kivshar, Dr Dragomir Neshev, said this observation offers “a cheap way to manipulate light on a chip”.

The Lawrence Berkeley National Laboratory boffins undoubtedly agree. In a nearly-identical announcement, the LBNL scientists state that they have discovered a way “to manipulate light at an extremely small scale beyond the diffraction limit”.

El Reg contacted the ANU’s Yuri Kivshar regarding the similarity of the two announcements. He described “bendable light” as a long-awaited breakthrough that has been pursued by not two, but four groups – the others in China and Korea – all of which have either achieved or are close to reporting similar results.

Both the Australian and US groups have submitted academic papers covering their work; the ANU’s results will appear in Physical Review Letters at the end of August, while the LBNL’s work is to be published in Optics Letters. Both groups also presented at the CLEO / QELS conference in Baltimore last May.

What happens to the light

Now for the science stuff, for which I thank Professor Kivshar for providing an advance copy of the Physical Review Letters paper.

The Airy beam was predicted 30 years ago. It’s a particular class of light that resists diffraction, and (to quote the paper) tends to “accelerate along parabolic trajectories”**. When small Airy beams are projected along the surfaces of particular materials, like graphene or magnetic films, they don’t spread with propagation: the beam at the end of its trip is far less diffracted than normal (Gaussian) light.

So how do you “bind” a light beam to a surface?

The trick – referred to by both the ANU and the LBNL – is in yet another property called the surface plasmon polariton, or SPP. In this “quasiparticle” (it’s a strange world, isn’t it?), photons are coupled to the free electron oscillations in metals.

This creates a “flatland” photonics: a planar environment in which the light beam stays within a wavelength of the surface across which it’s propagating.

It seems that the chance to yell “Eureka!” arrives if the beam that’s bound to the surface happens to be an Airy beam – because then, you get the double jackpot: a beam of light that stays where it’s put (on the surface of the chip) and can bend.

It’s early days, but both the ANU and LBNL identify managing and manipulating the “bendability” of Airy beams as the ultimate aim: that way, as Xiang Zhang, who leads the LBNL project, says: “This is promising not only for applications in reconfigurable optical interconnections but also for precisely manipulating particles on extremely small scales”.

Small wonder, then, that universities are in such intense competition over the Airy beam. ®

*Before the commenters pull me into line: El Reg is well aware that large gravitational fields, such as those around stars, can bend light. But this kind of light doesn’t need a star to bend it, which is a lot more convenient if you want to fabricate chips based on being able to bend the beam.

**And again: by "acceleration" we don't mean "the beam goes faster than the speed of light". If something travels in a curve, it is because it experiences acceleration in the direction of the curve - the ball you throw is accelerated towards Earth, for example. ®