Sunday, May 18, 2008

SPINTRONICS

The visionary who first thought of using the spin polarization of a single electron to encode a binary bit of information has never been identified conclusively. Folklore has it that Feynman mentioned this notion in casual conversations (circa 1985), but to this author’s knowledge there did not exist concrete schemes for implementing spintronic logic gates till the mid 1990s. Encoding information in spin may have certain advantages.
First, there is the possibility of lower power dissipation in switching logic gates. In charge based devices, such as metal oxide semiconductor field effect transistors, switching between logic 0 and logic 1 is accomplished by moving charges into and out of the transistor channel. Motion of charges is induced by creating a potential gradient (or electric field). The associated potential energy is ultimately dissipated as heat and irretrievably lost. In the case of spin, we do not have to move charges. In order to switch a bit from 0 to 1, or vice versa, we merely have to toggle the spin. This may require much less energy. Second, spin does not couple easily to stray electric fields (unless there is strong spin-orbit interaction in the host

material). Therefore, spin is likely to be relatively immune to noise. Finally, it is possible that spin devices may be faster. If we do not have to move electrons around, we will not be limited by the transit time of charges. Instead, we will be limited by the spin flip time, which could be smaller.

[get this widget]

0 comments:

search

Google