Designing for long distance travel: tens of µm

Artist’s impression of highly conductive graphene, in which electrons keep their magnetization, or spin (pink arrows) much longer than they do in ordinary conductors such as copper and aluminum. Credit: M Venkata Kamalakar et al/Nature Communications

It is all relative, of course, but when you listens someone talking about long distance travel you don't associate it with a few tens of µm distance!

Yet, this is what researchers at Chalmers University of Technology are doing: moving electrons for a few tens on µm, preserving their spin. And in this there is the catch.

Electron spin is quite fragile, you move the electron a few nm and the spin may change. We usually say that the spin is comparable to the rotation direction of a top: the electron rotates in one direction and you get an up spin, it rotates in the opposite direction and you get a down spin. Actually electrons don't rotate in the usual sense. They are both particles and waves but the scientists have decided to call one of their property spin because some of its characteristics make it looks like a top rotation.

Researchers have found ways of using this spin to store information: if it is up then you can associate the value 0, if it is down you associate the value 1. Hence spin is a good (potential) substrate for storage, and a very dense one, since theoretically one can squeeze what now requires a thousands atoms into a single atom.

This has given rise to spintronics, electronics at the atomic level.

When you store a bit you keep it in a very specific place, no need to move the electron around. However, if you want to process that bit that it has to be moved a few tens of µm through various logic gates in a chip. This is where spintronics fail (today). By moving the electron you change its spin (in a random way).

Researchers at Chalmers have found that graphene when used as a conductor (which it is, unless you tamper with it including different sorts of atoms here and there) has the property of preserving electron spin when the electron moves on the graphene layer. It becomes possible to move it to the place (on the layer) where there is a gate and where it can be processed. Of course, creating a gate on graphene requires creating a band gap (which is typical of semiconductors) and that requires altering the perfect hexagonal carbon structure with different atoms. 

We are still far from having an industrial use of graphene that can match (both technically and economically) silicon, but we are seeing interesting steps forward.

Author - Roberto Saracco

© 2010-2018 EIT Digital IVZW. All rights reserved. Legal notice