Discovery of the worlds smallest transistor - breaking the barrier

[glassjaw003]

A new article published in Science - http://science.sciencemag.org/content/354/6308/99

Phys.org article on the subject - http://phys.org/news/2016-10-materials-smallest-transistor-nanometer-carbon.html

By using MoSi2 instead of Si, they were able to produce a smaller  (~1 nm) gate without the electrons going out of control due to tunneling. Good news for all of us, this discovery could further the timeline of Moore's Law. Although, this result is just preliminary, not yet to the application stage.

As a side note, I am taking a class with Dr. Moon Kim. He is a really great guy.

[Coaxialgamer]

6 minutes ago, glassjaw003 said:

A new article published in Science - http://science.sciencemag.org/content/354/6308/99

Phys.org article on the subject - http://phys.org/news/2016-10-materials-smallest-transistor-nanometer-carbon.html

By using MoSi2 instead of Si, they were able to produce a smaller  (~1 nm) gate without the electrons going out of control due to tunneling. Good news for all of us, this discovery could further the timeline of Moore's Law. Although, this result is just preliminary, not yet to the application stage.

As a side note, I am taking a class with Dr. Moon Kim. He is a really great guy.

Further continue moore's law , yes , but at what price ? A 7nm SOC is already several times more expensive to produce than a 28nm one , and requires significantly more work ... 

What's the point of having smaller transistors for us enthusiasts if we get less of them ?

[KingUnicorn]

carbon nanotube trANSISTORS

[Curufinwe_wins]

Cool stuff but..

 

Oh god. I'm sure the actual article is less buzzfeedy but I swear these articles for the common folk are just so absolutely awful....

 

Proof of concept... this is a proof of concept of a proof of concept. 

 

Btw if I hear another one of these articles referencing quantum tunneling incorrectly again, some poor animal is going to be shot as a proxy.

 

NEWSFLASH! QUANTUM TUNNELING IS A NON-TRIVIAL ISSUE EVEN WITH 14NM TRANSITORS!

[Doobeedoo]

Exciting times seeing how much we can shrink. 2020+ will be very interesting to see where processors go. We will still be on silicon for some time now though but can't wait to see some quantum leap in progress with new materials and all.

[Sauron]

Moore's law is dead already I believe, the shrinking rate has slowed down. If we can reliably go down to 1nm on silicon compounds we probably still have about 10 years before we'll be forced to move to something completely different.

2 hours ago, Coaxialgamer said:

Further continue moore's law , yes , but at what price ? A 7nm SOC is already several times more expensive to produce than a 28nm one , and requires significantly more work ... 

What's the point of having smaller transistors for us enthusiasts if we get less of them ?

Intel has been throwing out the same old hyperthreaded quad cores for 6 years, let's be honest, their margins aren't small on the consumer lineup.

[Notional]

Scientist from Berkeley Lab's Materials Science Division has succeeded in creating a functional 1 nanometer transistor gate.

 

The team achieved this by using carbon nanotubes as gates inside of a material used as engine lubricant. With this breakthrough, they shatter the concern that sub 5 nm transistors were impossible due to quantum tunneling of the electrons, making them "jump" or leak onto other transistors, thus shorting out the chip.

Quote

Javey's team was able to hit the 1-nanometre limit by using carbon nanotubes with a material called molybdenum disulphide (MoS2), which is sometimes used as an engine lubricant.

 

To put this into perspective, Intel's latest Kaby Lake CPU's uses a 14nm FinFet process node:

 

This breakthrough could lead to Moore's law, that states that the number of transistors in a chip will double each year. As such we could see a massive jump in performance, and an ongoing performance improvement in the industry, that can carry us much further, until quantum computing becomes a feasible thing.

Quote

"This work is important to show that we are no longer limited to a 5-nanometre gate for our transistors," says Javey. "Moore's Law can continue a while longer by proper engineering of the semiconductor material and device architecture."

 

Source:

Article: http://www.sciencealert.com/scientists-just-developed-the-world-s-smallest-ever-transistor

Intel picture: http://www.intel.com/content/dam/www/public/us/en/images/illustrations/14nm-vs-22nm-metal-gate-rwd.png.rendition.intel.web.576.324.png

[TheRandomness]

Now this is cool (literally)

How about less than 1nm?

[Zyndo]

3 minutes ago, TheRandomness said:

Now this is cool (literally)

How about less than 1nm?

1nm has existed for all of a few minutes and its already not good enough for us enthusiasts eh? haha

[Notional]

1 minute ago, TheRandomness said:

Now this is cool (literally)

How about less than 1nm?

Then you are making a transistor composed of a split neutron, known as a quark. So that would make the transistor smaller than an atom. I doubt we will see that in our lifetime, if ever.

[Cla55ifi3xd]

3 minutes ago, Notional said:

Then you are making a transistor composed of a split neutron, known as a quark. So that would make the transistor smaller than an atom. I doubt we will see that in our lifetime, if ever.

The diameter of an atom ranges from about 0.1 to 0.5 nanometers (1 × 10^-10 m to 5 × 10^-10 m).

[Notional]

3 minutes ago, Cla55ifi3xd said:

The diameter of an atom ranges from about 0.1 to 0.5 nanometers (1 × 10^-10 m to 5 × 10^-10 m).

Fair enough, but you still have to factor in the distance between the atom parts where the current can run.

[Cla55ifi3xd]

4 minutes ago, Notional said:

Fair enough, but you still have to factor in the distance between the atom parts where the current can run.

We have already cracked that with  quantum computeing different I know but I love being contradictory. 

[Notional]

Just now, Cla55ifi3xd said:

We have already cracked that with  quantum computeing different I know but I love being contradictory. 

Well we are only limited by what we can do today. That limit shifts every day, so who knows where we will end up. However I think we will shift to quantum computing before 1nm transistors are obsolete.

[Dionyz]

why don't they just replace silicon, and make a carbon nanotube processor? 

[bob345]

Its one thing to make a functional 1nm transistor, but what about large scale production? If this process cant be used for large scale manufacturing then this technology will never see any kind of practical use. Little has been said in terms of production yields and cost. That is what really matters here. 

[MEC-777]

Topic already posted. 

 

 

[Coaxialgamer]

2 hours ago, Sauron said:

Moore's law is dead already I believe, the shrinking rate has slowed down. If we can reliably go down to 1nm on silicon compounds we probably still have about 10 years before we'll be forced to move to something completely different.

Intel has been throwing out the same old hyperthreaded quad cores for 6 years, let's be honest, their margins aren't small on the consumer lineup.

never said they were . But die space is getting more and more expensive , so they just make smaller chips ...

[Coaxialgamer]

50 minutes ago, MEC-777 said:

Topic already posted. 

 

 

that is this thread ...

[Curufinwe_wins]

10 minutes ago, Coaxialgamer said:

that is this thread ...

He merged them already that is why you get @Notional's really nice OP statement in the middle of this thread.

[patrickjp93]

5 hours ago, Dionyz said:

why don't they just replace silicon, and make a carbon nanotube processor? 

Intel and Samsung are both racing for it. IBM was too, but it's given up. It would be completely new techniques. Research and testing has to be done, and known techniques need to be stretched to make investments last.

[Cla55ifi3xd]

5 hours ago, Dionyz said:

why don't they just replace silicon, and make a carbon nanotube processor? 

Because for some reason they can't make a nano tube that leaves a laboratory lol. 

[Prysin]

6 hours ago, Notional said:

Then you are making a transistor composed of a split neutron, known as a quark. So that would make the transistor smaller than an atom. I doubt we will see that in our lifetime, if ever.

if you want to see something almost that small. well. you know where to look.

[Notional]

1 minute ago, Prysin said:

if you want to see something almost that small. well. you know where to look.

Yeah, but the court order states that you are not allowed to take off your pants in public anymore.

[Prysin]

2 minutes ago, Notional said:

Yeah, but the court order states that you are not allowed to take off your pants in public anymore.

shame isnt it.