Print 13 comment(s) - last by ramprun.. on Jun 27 at 12:17 AM

Device operates at room temperature; feature with is currently 20 nm, could be shrunk smaller

Michigan Technological University (Michigan Tech/MTU) physics professor Yoke Khin Yap is concerned that the current route of semiconductor development is untenable for extending Moore's Law.  He comments [press release], "At the rate the current technology is progressing, in 10 or 20 years, they won’t be able to get any smaller.  Also, semiconductors have another disadvantage: they waste a lot of energy in the form of heat."

He believed the solution to this challenge lay in new, novel materials.

He began testing designs using tiny globs of metal called "quantum dots" (QDs) sprinkled on a nanoinsulator.  For the insulator substrate he chose boron nitride nanotubes, known as BNNTs.  For the quantum dots he used gold, an ideal material for making regular, precisely-sized QDs.

The team needed a way to position the dots in atomic space on the nanotube, so they turned to using a laser.  Using this method, they were able to positions gold QDs that were a mere 3 nanometers in diameter -- or roughly 1/7th the size of transistors produced at current circuit manufacturing nodes.

BNNTAn artist's rendering of the nanotube transistor [Image Source: MTU]

Testing the design in collaboration with Oak Ridge National Laboratory (ORNL), they hooked an electrode up to each end of the construct and tested it at room temperature.  They observed quantum tunneling -- the hallmark phenomena necessary to construct a non-semiconductor transistor.  Electrons "jump" (or tunnel) from one gold QD to the next, as current is applied.

The researchers were able to control the voltage to switch this conduction on and off, forming a transistor.  Professor Yap describes, "Imagine that the nanotubes are a river, with an electrode on each bank. Now imagine some very tiny stepping stones across the river.  The electrons hopped between the gold stepping stones. The stones are so small, you can only get one electron on the stone at a time. Every electron is passing the same way, so the device is always stable."

Past transistors made from materials other than semiconductor typically had to be cooled with liquid helium to operate well; by contrast Professor Yap's design performs well at room temperature.

Currently each nanotube is 1 micron long and 20 nm wide -- making these transistors on par with current designs.  But the researchers expect these transistors to scale better than semiconductor sizes as tube diameter and lengths are shrunk.  The team already has established methods to deposit aligned substrate "carpets" so now all that remains is developing methods to mass-position the nanodots (as individual laser positioning is prohibitively slow for making the billions of transistors in a modern IC).

BNNT carpet
Electron micrographs of nanotube "carpets" grown by Prof. Yap's team back in 2011.
[Image Source: MTU]

Professor Yap, who has filed for a patent on the design and manufacturing process, comments, "Theoretically, these tunneling channels can be miniaturized into virtually zero dimension when the distance between electrodes is reduced to a small fraction of a micron."

The best feature of the transistors is that there's no electrons (according to the authors) lost between gold nanodot and no gold nanodot -- a heat generating phenomena known as "leakage". By contrast, leakage is a massive problem for nanoscale silicon-based transistors, limiting clock speeds and circuit density.

In addition to the patent his work was published [abstract] in a peer-reviewed journal article in the Advanced Materials journal from publisher Wiley.

Sources: Advanced Materials [abstract], MTU [press release]

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

HEat is one issue...
By Motoman on 6/25/2013 3:38:24 PM , Rating: 5
...but IIRC the biggest problem with the current silicon tech is electron migration. As you make everything smaller and smaller, it's harder to keep the electrons where you want them - they can "migrate" out on their own and just give you the finger, leaving you with a device that doesn't really work well.

Kinda sounds like this solves that issue too. Somehow. Not sure how, but hey, gold and nanotubes. Whee!

RE: HEat is one issue...
By Odysseus145 on 6/25/2013 4:09:56 PM , Rating: 2
I wonder how close you can put two parallel devices before the electrons jump between transistors.

RE: HEat is one issue...
By DerMack on 6/25/2013 6:13:58 PM , Rating: 2
you're mixing up things there a bit

electromigration - transport of material caused by the gradual movement of the ions in a conductor due to the momentum transfer between conducting electrons and diffusing metal atoms.
so the metal in the conductor traces and stuff moves about beacause of high current densities, causing open circuits and/or shorts (the redeposited metal forms hillocks and stuff that can make contact with nearby traces or physically crack things open). And as things get smaller the current densities tend to go up, and of course a lot less metal needs to be moved for the trace to fail.

what you're talking about sounds more like tunneling. :)

RE: HEat is one issue...
By Motoman on 6/25/2013 9:16:47 PM , Rating: 2
No, although the 2 terms are similar, they're very different. Electromigration, as you've noted, is actually a change in the material itself - grossly, like you broke a chunk of silicon off and moved it around inside the semiconductor.

Electron migration in general refers to the movement of electrons from one location to another, based on understood particle physics. In and of itself, "electron migration" isn't a bad thing (and actually, generally speaking, is a good thing...electrically speaking)...and perhaps I should have been more clear originally, but with reference to semiconductor designs, "electron migration" is brought up as a problem in that the electrons find places to migrate to that you didn't intend for them to go.

RE: HEat is one issue...
By BRB29 on 6/26/2013 8:11:23 AM , Rating: 3
Electromigration is movement of atoms caused by momentum of transfer of electrons(current and wind force). I don't normally hear people use electron migration, they prefer using "current".

RE: HEat is one issue...
By Motoman on 6/26/2013 11:08:44 AM , Rating: 2
If you google up the term, you find many papers where the phrase is used, especially in relation to semicondutors.

While technically the same thing, "electron migration" seems to get used when you're trying to refer to a singular movement of an electron in a specific instance, as opposed to "current" which of course is the intended large-scale flow of electrons through a given conductive material.

RE: HEat is one issue...
By boobo on 6/25/2013 10:04:57 PM , Rating: 3
The way they solve it is by moving (through quantum tunneling) only one electron from gold dot to gold dot, instead of pushing a current of electrons, like semiconductors do.

When you juice up the electrode on the source side, you're not pushing the electrode's current through the path of nanodots. You are determining how many hops a single electron will reach, so the very concept of leakage does not apply.

You sure about that?
By bug77 on 6/25/2013 3:46:07 PM , Rating: 3
...quantum tunneling -- the hallmark phenomena necessary to construct a transistor

I'm pretty sure current transistor are doing just fine without quantum tunnelling.

RE: You sure about that?
By Odysseus145 on 6/25/2013 4:07:40 PM , Rating: 2
It was poorly worded, but I think the implication was that tunneling is necessary for non-semiconductor transistors.

RE: You sure about that?
Multiple Laws
By Shig on 6/26/2013 2:07:17 PM , Rating: 3
It's true that Moore's Law is slowing down. We'll be going to half node shrinks when we start going below 20-22nm.

Koomey's Law is says that performance per watt increases similar to a Moore's Law progression. Over the past few years with mobile growing so fast, this law is actually speeding up to compensate for Moore's Law slowing down.

Umm Hmmm
By R!TTER on 6/25/2013 4:10:50 PM , Rating: 2
"To be or not to be" a semiconductor "that is the million dollar question: (:

By ramprun on 6/27/2013 12:17:00 AM , Rating: 2
Pretty sure quantum dots are semiconductor

"This is about the Internet.  Everything on the Internet is encrypted. This is not a BlackBerry-only issue. If they can't deal with the Internet, they should shut it off." -- RIM co-CEO Michael Lazaridis

Most Popular ArticlesAre you ready for this ? HyperDrive Aircraft
September 24, 2016, 9:29 AM
Leaked – Samsung S8 is a Dream and a Dream 2
September 25, 2016, 8:00 AM
Yahoo Hacked - Change Your Passwords and Security Info ASAP!
September 23, 2016, 5:45 AM
A is for Apples
September 23, 2016, 5:32 AM
Walmart may get "Robot Shopping Carts?"
September 17, 2016, 6:01 AM

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki