Single Atom Transistor

[Reynod]

http://finance.ninemsn.com.au/technology/8422385/single-atom-transistor-busts-the-records

Researchers in Australia said on Sunday they had made with pinpoint accuracy a working transistor consisting of a single atom, marking a major stride towards next-generation computing.

The device comprises a single phosphorus atom, etched into a silicon bed, with "gates" to control electrical flow and metallic contacts that are also on the atomic scale.

"Our group has proved that it is really possible to position one phosphorus atom in a silicon environment, exactly as we need it, with near-atomic precision, and at the same time (incorporate) gates," said lead scientist Martin Fuechsle.

 

[hunter315]

Next step, being able to automate the process and do it consistently then we are talking, it probably has amazing thermal efficiency, but would probably take them a decade to make the equivalent of an 8086 so still a long ways to go, but still a promising development.

 

[Reynod]

I wonder what order of magnitude current lithographic process would need to improve?

 

[wanamingo]

Being able to manipulate anything at an atomic level is mindblowing. Then actually making it work would have been impossible even 10 years ago.

Supercooling might keep it out of the hands of most of the public though.....My liquid helium reserves are quite low.

 

[riser]

The Moore's Law set the date for 2020 that this technology would be achieved. They're trying to get ahead of the times.. and I'm betting 2020 will be the eventual date of mass production. Moore's law considered more than what the basic idea consists of today. For example, economic conditions were factored in among other things. With the way things are today and the world economy on the fringe, 2020 might very well be the realistic date.

Interesting stuff though.

 

[dogman_1234]

I am thinking of some issues currently.

First off, the amount of electrons passing through the P atom would be troublesome. can't magine a stable atom with 1x10^12e electron passing through it every second.

Second: Tunneling. QTE would have an impact on the transistor itself.

Thirdly, what about a fabrication process. There is no direct fabrication process for even current nanotechnology...except photo-lithography. A still ancient form of etching.

What I see is for this to happen for mass production is we need to develop a device that displaces atoms over a sheet of wells. These wells act as a template and will switch 'gates' around the transistor, enabling a possible 3 state system.

 

[riser]

People over estimate what they can do in a year and under estimate what they can do in 10.