Intel works with Honeywell to make new Materials for 7 nm process

[Tylandord]

So I watched the wan show on Youtube about the new materials for Intel's future plans for the 7 nm process. So I talked to my source who works over at the Honeywell factory in Spokane Washington, and I told him about the possibilities of III-V(indium-gallium-arsenide), Graphene, and carbon nano-tubes possibly being candidates for the new materials in Intel's processors instead of silicon. Then earlier today they told me that there was a huge worldwide Honeywell meeting being broadcast live to all of the Honeywell people and they talked about issues on this and that. Then they asked "Does anyone have any questions," but because the people talking were real big shots everyone was silent. So my source decided to stand up and ask "I have a question" to the whole world of Honeywell people.

And they asked "So what are we going to do when we can't use silicon anymore, and need to start manufacturing new materials like indium-gallium-arsenide".

the big shot guy said "That is a very good question." he continued "You're talking about Intel's new 7 nm process aren't you?"

my source replied "I am indeed".

Then big-shot guy said "I will be pleased to tell you that Intel will be working with with Honeywell to manufacture these new materials to make their 7nm processors".

So the short of it is Intel is already working towards manufacturing these new 7 nm processors with Honeywell. Personally, I'm stoked about the new possibilities for new materials in future technologies. This is super cool.

[Tonylu1595]

Graphene. Always Graphene.

[Hatsune Miku 「 」]

1st. Indium Gallium Arsenide

[Tylandord]

I agree, Graphene seems to be the most interesting material in my mind. Although I don't know much about III-V materials. I should do some research.

[Peter Vaughan Truslow]

from what I know of graphene, the lattice is self-repairing from neighboring free carbon, which could prove problematic in the real world, but I don't think silicon will hold out much longer.

Carbon nanotubes are made of graphene rolled at a particular angle, so my conclusion is InGaAs will provide some of the answers we need.

 

from the wikipedia page on InGaAs, 

HEMT devices using InGaAs channels are one of the fastest types of transistor

http://en.wikipedia.org/wiki/Indium_gallium_arsenide#Transistors

it looks like InGaAs might once again allow us to push our clocks higher as we haven't in so long

edit: but we're not there yet, the smallest transistor we have from InGaAs is a 22nm MOSFET. it's the smallest thing they've done without silicon.

[Real_PhillBert]

I default to those more knowledgeable than I on these matters. @LukaP is far more educated on this than I.

[Tylandord]

I understand your point and I agree. III-V (indium-gallium-arsenide) does seem the most promising, I'm just a Graphene fanboy.

[Tylandord]

from what I know of graphene, the lattice is self-repairing from neighboring free carbon, which could prove problematic in the real world, but I don't think silicon will hold out much longer.

Carbon nanotubes are made of graphene rolled at a particular angle, so my conclusion is InGaAs will provide some of the answers we need.

 

from the wikipedia page on InGaAs, 

HEMT devices using InGaAs channels are one of the fastest types of transistor

http://en.wikipedia.org/wiki/Indium_gallium_arsenide#Transistors

it looks like InGaAs might once again allow us to push our clocks higher as we haven't in so long

edit: but we're not there yet, the smallest transistor we have from InGaAs is a 22nm MOSFET. it's the smallest thing they've done without silicon.

I agree that III-V is probably a better solution. I'm really just a graphene fanboy.

[wpirobotbuilder]

We already knew they were looking for new materials.

 

Also, @Tylandord, you really shouldn't reveal how these details got leaked to you, because they are almost certainly confidential information.

[Blackie Sheen]

I agree that III-V is probably a better solution. I'm really just a graphene fanboy.

 

Graphene fanboy. LOL. That exists...

 

[Lazmarr]

-snip-

Graphene, material to have the least impact on health and isn't a potent toxin to humans...think of the lives you could save by going with graphene instead of an arsenic-based alloy

 

[Tylandord]

@wpirobotbuilder I'm not really worried about that, they haven't said anything besides they are looking into new materials, which we already knew. And plus they haven't even said what materials will be used. Also it's not like I leaked the specs or got a photo of the new processor. Honeywell manufactures a very large portion of materials for Intel and other computer parts manufacturers already so it should be no surprise that they would continue to work together.

[Tylandord]

@Lazmarr I totally agree on the point of graphene being better for the environment and to the people that are exposed to it. It's just a safer option.

[szomeret]

Hoping for silicon. Ga-Ar is the most likely though.

[han_han08]

just got lost in Intel's website a few days ago. Will these materials be Conflict-free? 

[Tylandord]

@szomeret They are already using silicon i just put it in as a joke.

[szomeret]

@szomeret They are already using silicon i just put it in as a joke.

I'm hoping they won't have to switch to new materials. They're all more expensive than silicon...

[Tylandord]

I'm hoping they won't have to switch to new materials. They're all more expensive than silicon...

Intel is getting to the point where they cannot continue to use silicon because its becoming more and more difficult to manufacture, ergo higher manufacturing costs. In the long run it will be cheaper and more efficient to switch materials.

[AxelRantila]

I'm hoping they won't have to switch to new materials. They're all more expensive than silicon...

Isn't more or less all  fabs having issues already at 16nm? If Intel can find something new it might actually save lots of costs long-term wise

[szomeret]

Intel is getting to the point where they cannot continue to use silicon because its becoming more and more difficult to manufacture, ergo higher manufacturing costs. In the long run it will be cheaper and more efficient to switch materials.

Samsung said they'll be able to scale on silicon to 7nm. Yes, in the long run it will be necessary but the cost savings are to stay at silicon for quite some time. I suspect Intel is going to transition before the pack (at 7nm rather than after) due to its much higher profit margins allowing more expensive leading-edge technology.

Fabs have been having issues for ever. There is always more to mitigate.

[Tylandord]

Isn't more or less all  fabs having issues already at 16nm? If Intel can find something new it might actually save lots of costs long-term wise

Exactly. I couldn't agree more.

[patrickjp93]

Isn't more or less all  fabs having issues already at 16nm? If Intel can find something new it might actually save lots of costs long-term wise

They're having trouble with 20nm actually, and more specifically mastering FinFET processes at that feature size). The 16/14nm naming schemes for TSMC and Samsung/GloFo are just marketing BS.

 

https://www.semiwiki.com/forum/content/3884-who-will-lead-10nm.html

[Nielips]

Graphene, material to have the least impact on health and isn't a potent toxin to humans...think of the lives you could save by going with graphene instead of an arsenic-based alloy

This is not true, for one most nano materials have an unknown health impact on humans, but also CNT's of a certain size are known to have the same toxicity mode as asbestos.

[Nielips]

This is not true, for one most nano materials have an unknown health impact on humans, but also CNT's of a certain size are known to have the same toxicity mode as asbestos.

On top of that the toxicity of arsenic is highly dependent on the for it is in. Furthermore the toxicity of everything is completely dependent on the dose and mode in which you are exposed to it.

[Bsmith]

ohh mighty @LukaP educate us on the best material choice!