Caltech sends light on a one-way trip
Optical computing gets a small step closer
A group of Caltech researchers is claiming what it describes as a key breakthrough in photonic computing: an optical diode using linear, rather than non-linear, materials.
The diode – or rather, components that mimic the function of the diode in the optical domain – is important to optical computing, because it helps isolate signals from each other. For example, it protects signals from interference from backscattered noise.
According to PhD scholar Liang Feng, the optical isolator – a device that allows light to travel in one direction only – is “something that scientists have been pursuing for 20 years”.
One way to isolate light is with magnets, but this isn’t feasible at the microscopic scale needed to build integrated optical chips capable of carrying out computing functions: the magnets need to be too big. The other is to use non-linear optical materials. This works well, but the researchers say non-linear materials are difficult to integrate with traditional silicon-based fabrication methods.
In their proof-of-concept paper, Non-reciprocal light propagation in a silicon photonic circuit, published in Science, the researchers say they have transmitted light in a symmetric mode in one direction, but in the other direction, the light changes to asymmetric mode. The paper is outlined here.
This isn’t actually “blocking” the light in the return direction – not in the way that a diode blocks an electrical signal in the “wrong” direction – but it means that the two optical signals don’t interact with each other.
The Caltech group built an 800nm x 200nm waveguide, and added materials with different reflective and refractive properties to break the symmetry of the light passing through the waveguide.
The next step for the researchers will be to take the proof-of-concept and create optical isolators suitable for integration onto integrated circuits. ®
@one way mirrors
Actually, most one way mirrors work considerably different. They just are mirrors with a bit of transmission going on. So essentially they are slightly transparent mirrors. If the other side is much darker than the side you are on, they look the same as normal mirrors. However if you had a high dynamic range camera and would simply substract the much brighter reflection, you'd see through the mirror to the other side.
When will we see the first...
... full-fledged, Star-Trekkin...-Across-the-Universe, Isolinear Optical Chip?
At first i thought "wow thats clever" then i got all confused when i realised that one way mirrors / windows have been doing the same thing for a lot longer.
I realise that process is probably completly different but hmmmmmm physics has never been my strong point :)