World’s Smallest Transistor

H

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
Researchers have built the world's smallest transistor out of a material they say could one day replace silicon. How small is the new transistor? Ten atoms wide and only one atom thick.

The transistor, essentially an on/off switch, has been made using graphene, a two-dimensional material first discovered only four years ago. Graphene is a single layer of graphite, which is found in the humble pencil.
Click to expand...
 
Eulogy said:
Two dimensional material? Someone tell me why I can't quite wrap my head around that ;).
Click to expand...

Well, if your only 1 atom thick you can only have electrical conduction and heat tranfer in the dimensions that are multiple edges thick, i.e. 2 dimensions.
 
Eulogy said:
Two dimensional material? Someone tell me why I can't quite wrap my head around that ;).
Click to expand...

This'll help ( I think): essentially two dimensional, as it's only one atom thick, so the third dimension is of sub-atomic scale.
 
It's 3 dimensional. It's impossible to have a physical object only be 2 dimensions. It doesn't matter if it's "only" one atom thick... sure, you can't get any thinner, but you still have a thickness!

...not that it matters at all. I'm just nit-picking.
 
Eulogy said:
It's 3 dimensional. It's impossible to have a physical object only be 2 dimensions. It doesn't matter if it's "only" one atom thick... sure, you can't get any thinner, but you still have a thickness!
Click to expand...

Agreed.
 
50% yield per transistor ...so the odds of getting a defect-free billion transistor processor? They can depend on luck when cutting a single transistor in a lab, but in order for this to become practical somebody has to figure out how to fabricate a 1 atom thick layer of graphite-aligned carbon on a [very flat] substrate the size of a die, if not a wafer. Then they figure out how to cut transistors into it.

Another way to fabricate at these scales is nano-assembly, which may actually be a smarter way to do it. That process [in theory] consists of placing individual atoms to construct the circuits. Incidentally, when your transistors are 1 atom thick and 10 atoms wide, building them 1 atom at a time doesn't seem so unreasonable.

Suffice to say it will be a while before this makes it from the lab to commercial applications, but the potential wins are so huge the material is definitely getting lots of [rightfully deserved] attention.
 
While this article is an interesting read, and a possible look into the future of what might be our way of life, I would liken it to the proverbial putting the cart in front of the horse. The entire premise, at least to my understanding, appears to be more about the high tech brew of gadgets and gizmos that we can come up with in darkly lit laboratories instead of what can be made practicable to assembly lines in an economically feasible way so that it can be available to people like you and me.
 
That's usually how things like this start out. Someone makes one in a lab and shows that it can be done. Then they attempt to mass produce it economically. Sometimes it can be done, sometimes it takes 10,20,30 year for someone to figure out how and sometimes it's just turns out to be a laboratory curiosity; because no one can figure out how to make it at a reasonable cost on a large scale. Aluminum was like that once upon a time. At one time it was more valuable than gold and only impure forms could be made, at great expense, in a lab.
 
This is som pretty cool stuff. Years off, if not fully impractical.
We'll see III-Vs on Si soon, and Trigate FETs but i don't think we'll be seeing graphene for a very very long time.
 
What about Silicon Germanium? I remember they had a transistor clocked at like 350ghz at ambient air temps.
 
No one had a use for the laser when it was first invented and now is used everywhere from playing cd's to tooth whitening. I am waiting for the day they announce a 3d transistor that works without the constraints of a 2d device.
 
my first thought (no pun intended) was actually of brain chemistry mimicking... technically speaking, the artificial intelligence proponents are going to like this.. the programming would take ages though...

the amount of energy needed to power one of these is so small that if grafted on to a human that produces chemical, friction, and thermal energy, you could map the genome during 10 min on an elliptical.. presuming storage could keep up.

very powerful computational tools to come..
 
well all it has to do is pass a signal through right?, and either stop it or let it continue. so if it were just a string of atoms, it would work right.

Is it like a rigid molecule?
 
s_s256 said:
It is graphite only a single atom deep. Apparently it looks like honeycomb under a microscope.

http://en.wikipedia.org/wiki/Graphene
Click to expand...


thats not under a microscope :) its just a 3d representation of the crystal lattice structure of the molucule.

any carbon-only material (ie diamond) would look pretty much the same, only it would be many lairs thicker.

http://www.chm.bris.ac.uk/webprojects2001/andrews/carbonforms.jpg
pretty good pic there...it even has a carbon nano-tube pictured.
 
You must log in or register to reply here.
Back
Top