Researchers Create Single-Atom Transistor

H

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
For those of you that haven't seen this yet, Researchers from Purdue, the University of Melbourne and the University of New South Wales have created a working transistor from a single atom. Thanks to Robert for the link.

We have fabricated a single-atom transistor in which a single phosphorus atom is positioned between highly doped source and drain leads with a lateral spatial accuracy of ±1 atomic lattice spacing. We demonstrate that we are able to register source, drain and gate contacts to the individual donor atom and observe well-controlled transitions for 0, 1 and 2 electron states, in agreement with atomistic modeling of the device.
Click to expand...
 
This is pretty awesome news, but I want to know how far away this is from being put into practical application.
 
Bring on quantum computing. This is another step toward that end. We might just reach the technological singularity within my lifetime and the prospects of that are just incredible.
 
blitzkommando said:
Bring on quantum computing. This is another step toward that end. We might just reach the technological singularity within my lifetime and the prospects of that are just incredible.
Click to expand...

we are MORE then on track for that the up tick happens the end of this year XD
 
How small is that compared to the 28nm or whatever CPU size were on now?
 
Inu said:
How small is that compared to the 28nm or whatever CPU size were on now?
Click to expand...

A quick Google search tells me that a phosphorus atom is about 196 picometers (pm). A picometer is a trillionth of a meter...versus a nanometer which is a billionth of a meter.
 
Now can they turn this into a processor or memory? Chances are, we probably won't see this for decades.
 
Ashbringer said:
Now can they turn this into a processor or memory? Chances are, we probably won't see this for decades.
Click to expand...

It took 33 years to get from 3.0 um to 32 nm, 17 years to get from 2.0 um to .18 um, etc., so yeah, this won't be commercial for a while, but then again they usually have workable prototypes ten years in advance, so that is expected.
 
That's wonderful and cool and all, and I don't mean to trivialize the accomplishment, but where could they go from here? We find smaller, faster, and cooler (less heat-producing) ways of processing every couple of months, right? The only lighter element in the same class as phosphorus is nitrogen, which is a gas at standard temperature and pressure. So I doubt that would work. I don't see them using the nucleii of atoms either, due to the energy required there. Will this be a brick wall down the road?
 
Maxx said:
It took 33 years to get from 3.0 um to 32 nm, 17 years to get from 2.0 um to .18 um, etc., so yeah, this won't be commercial for a while, but then again they usually have workable prototypes ten years in advance, so that is expected.
Click to expand...

The objective of this, if i recall the article, was that they wanted to beat Moore's Law (which they did). And even if this isn't something usable immediately, it's still a huge step forward for quantum computing.
 
Eisenblut said:
Like fusion power. At any given point in time, it's 50 years away.
Click to expand...

more like 5 to 10 atm its now purely a mater of building the plants and putting the money in to the tech
just like going to the moon if we actually threw the weight of the US in to it it could be ready to go in less then 5
but no were wasting time on "green" energy that really isnt
 
Elios said:
more like 5 to 10 atm its now purely a mater of building the plants and putting the money in to the tech
just like going to the moon if we actually threw the weight of the US in to it it could be ready to go in less then 5
but no were wasting time on "green" energy that really isnt
Click to expand...

You are comparing a consumer product versus a large scale government funded project. This won't get government attention and for good reason, they have no reason to be in this side of technology.

They can move to be as green as much as they want but the amount of money our government is putting twords "green" stuff isn't actually very much at all. The biggest thing is to get all large corporations involved to PUSH it faster, at the same time you are also asking these companies to skip essentially all generations to very last one.

As much as you want it, its a very very big toll to ask of any company to essentially push this emmediately. Then you are stuck with the future development of technology past this, meaning we will be stuck with this technology for a long while building animosity amongst the rest of the technology sector for lack of improvement.
 
Eisenblut said:
Like fusion power. At any given point in time, it's 50 years away.
Click to expand...

no sorry e-cat will be here within the year. LENR is real I've seen it with my own eyes at the NASA Langley Research Center in Virginia.
 
Also, how did they get three electric states? 0 off, 1 on, but what's the third for?
 
Imaulle said:
no sorry e-cat will be here within the year. LENR is real I've seen it with my own eyes at the NASA Langley Research Center in Virginia.
Click to expand...

you left out the stuff going on it the EU with DEMO and ITER and TEST
 
Inu said:
How small is that compared to the 28nm or whatever CPU size were on now?
Click to expand...

eggrock said:
Depends on the atom in question.

According to this article, Intel's 22nm fab is < 100 atoms.
Click to expand...
In response to both of these, atom diameter/radius is usually measured in Angstroms which is 0.1 nm. Figure the diameter to be about 2 Angstroms or 0.2 nm

satsunada said:
The objective of this, if i recall the article, was that they wanted to beat Moore's Law (which they did).
Click to expand...
It depends on which version of Moore's Law you use. If you use the original iteration that Moore wrote, they have to be able to actually put some number of transistors on a chip without running into severe yield problems. Simply put, they have to be able to put a bunch of transistors on a chip relatively inexpensively.
 
Imaulle said:
sorry not familiar with any of that stuff. I've only seen the working NASA replica of rossis device
Click to expand...

ITER is being built right now and will out put 5x more energy then breakeven
DEMO will do 20x and will be hooked to the power grid wile ITER will just dissipate the heat
DEMO should be fully online in 2040
 
2040 what the fuck that is too long????? I don't think the human race will survive that long
 
satsunada said:
The objective of this, if i recall the article, was that they wanted to beat Moore's Law (which they did). And even if this isn't something usable immediately, it's still a huge step forward for quantum computing.
Click to expand...

You mean they have cheaply integrated a large number of them into an integrated circuit? Because I don't think they have cheaply integrated a large number of them into an integrated circuit.
 
Eisenblut said:
Like fusion power. At any given point in time, it's 50 years away.
Click to expand...

lol I imagine the dinosaurs experimenting with cold fusion (almost, just 50 years and we got it!). then they blew up their research facility and went extinct.
 
You must log in or register to reply here.
Back
Top