Caltech engineers create photonic 'diode'

Engineers claim to have reached a breakthrough in the field of photonic computing, bringing the technology - which replaces the electronic signals in a computer chip with light - a step closer to being a commercial reality.

A team of engineers led by the California Institute of Technology have published a paper in the journal Science which describes a new technique for isolating light signals on a silicon chip, which the team claims clears one of the major hurdles preventing the technology from hitting the big time.

The paper's method details the photonic equivalent to the electronic diode, a component which only allows signals to travel in a single direction in order to prevent interference. While that's been easy to achieve in electronics, the identical properties of a reflected light beam means that it's been impossible to isolate in the same way.

"We want to build something where you can see me, but I can't see you," the paper's lead author Liang Feng of Caltech's nanofabrication group explained. "That means there's no signal from your side to me. The device on my side is isolated - it won't be affected by my surroundings, so the functionality of my device will be stable. This is something scientists have been pursuing for 20 years."

The 'photonic diode' created by the team takes the form of a new type of optical waveguide measuring 0.8 microns wide. Light travelling in one direction passes through unaffected, but any coming from the opposite direction flips from a symmetric mode to an asymmetric mode which is unable to interact with the signal-bearing light.

It's not a new technology: nearly a century ago, magnetic fields were being used to change the polarisation of light for much the same reason, and around 50 years ago an alternative method using non-linear optical materials was developed. Both are unsuited to use in photonic computing, Feng claims: "You can't put a large magnetic field next to a computer. It's not healthy."

The use of non-linear materials would also mean throwing out decades of semiconductor manufacturing technology and starting from scratch. Silicon, which is used as the basis for the vast majority of semiconductors created today, is a linear material; a shift to a non-linear material would require a radical rethink of production methods.

The team's aim is nothing short of a complete replacement for electronics. "We want to take everything on an electronic chip and reproduce it on a photonic chip," Feng claims. That means a chip which is capable of general purpose computing in the same way as today's electronic circuitry, but potentially at a lower power draw and running at a far faster speed than is achievable with electronic circuits.

Sadly, that lofty goal is somewhat distant. Currently, the team has proven that an optical isolator constructed from linear materials can work, but has yet to create a version which can be integrated onto a silicon chip.

Although work is progressing, Feng claims, no time scale has been suggested for when this - rather more important - breakthrough may be achieved.