Self-driving cars are becoming better and more reliable. There are a number of hurdles to be overcome before they can possibly soon be completely autonomous on the roads. Above all, the lightning-fast detection of the environment and the recognition of people and obstacles still pushes today’s technologies to their limits.
Scientists led by Jürg Leuthold from the Institute for Electromagnetic Fields at ETH Zurich, together with colleagues from the National Institute of Standards and Technology (NIST) in the USA and Chalmers University in Gothenburg (Sweden), have now developed a new type of electro-optomechanical switch that can perhaps be used to elegantly solve both problems.
The magic agent used by the researchers is called plasmonics. In this technique, light waves are forced into structures that are much smaller than the wavelength of the light – which is not possible according to the laws of optics. This is however made possible by conducting the light along the boundary surface between a metal and a dielectric, i.e. an electrically weak or non-conductive substance such as air or glass. The electromagnetic waves of the light partially penetrate into the metal and excite electrons to vibrate, resulting in a hermaphrodite being of light wave and electronic excitation – the plasmon.
More than ten years ago, renowned physicists predicted that optical switches based on plasmons could herald a revolution in data transmission and processing, since both are much faster with photons than with conventional electronics. So far, however, commercial applications have failed due to the large losses caused by the transport of photons through plasmon components and the high switching voltages required.
The researchers now believe that these problems have been solved. At the heart of the electro-optomechanical switch that has now been developed is a gold membrane that is only 40 nanometers thin and a few micrometers wide, separated from a silicon substrate by an aluminum oxide disk. The size of the gap between the gold membrane and the substrate can be mechanically controlled by electrical forces. If a voltage is applied, the membrane bends slightly and the gap becomes smaller.
The size of the gap in turn determines whether a light wave simply flies on straight or is deflected around the gold membrane. This is where plasmons come into play: for a certain gap width, only plasmons with a certain wavelength can be excited on the gold membrane. If the light has a different wavelength, it is not coupled to the membrane and propagates in a straight line in the silicon waveguide.
“We have considerably lower losses than in previous electro-optical switches,” explains postdoctoral fellow Christian Haffner, who led the project as the first author of the recently published science article. “We have also made the gold membrane very small and thin so that we can switch very quickly and with low voltage.” The scientists have already been able to show that their new switch can be turned on and off several million times a second with an electrical voltage of just over one volt.
This eliminates the need for bulky and power-guzzling amplifiers previously used for electro-optical switches. In the future, the researchers want to further improve their switch by making the gap between gold and silicon even smaller. This will significantly reduce both the light losses and the control voltage.
There is plenty of application potential for the new switch. Lidar systems for automated cars, for example, where the intensity and direction of propagation of light beams have to be changed extremely quickly, could benefit from the fast and compact switches. And pattern recognition required to control the vehicles can be made faster with such switches.
To this end, the switches could be used in optical neural networks that are modelled on the human brain. There they would then be used as weighting elements with which the network “learns” to recognize certain objects.
Optical implementations of circuits which normally work with electric current, are also a hot issue in other areas. For the realization of quantum computers, for example, optical quantum circuits are intensively researched. So far, optical quantum circuits have been supported by classical optical switches.
However, this works only slowly and in the long run is incompatible with the low temperatures at which other quantum elements usually function. A fast switch that practically doesn’t heat up at all – like the one developed by Swiss researchers – could therefore be most welcome for such applications.
Source: eenewseurope
