No additional cost, only the magic

[Li Yisuo]

As discussed in https://www.semiwiki.com/forum/content/4091-intel-has-another-first-14nm-production.html, we learned without magic we have to bear additional cost. Jumping to the magic concept, I just found the last airbender uses his magic expertise to fly. Cool, isn't it?

So let's deal with it. What magic do we have today? GRAPHENE. Of course, 2D material, like wing the airbender carries above, which can support heavy stuffs on this top with just a mono atomic layer.
Here we go, instead of using additional mask, we do blanket etch. And then put our magic 2D material on top and then cover with low k material. Voila, the air gap is pure gap. RC reduction should be >50%. However our magic material is conducting. So in the end, we have to burn it off. Don't worry, it is very reactive. Any questions?

 

[Daniel Nenni]

What materials will be used for 7nm?

As discussed in https://www.semiwiki.com/forum/content/4091-intel-has-another-first-14nm-production.html, we learned without magic we have to bear additional cost. Jumping to the magic concept, I just found the last airbender uses his magic expertise to fly. Cool, isn't it?
So let's deal with it. What magic do we have today? GRAPHENE. Of course, 2D material, like wing the airbender carries above, which can support heavy stuffs on this top with just a mono atomic layer.
Here we go, instead of using additional mask, we do blanket etch. And then put our magic 2D material on top and then cover with low k material. Voila, the air gap is pure gap. RC reduction should be >50%. However our magic material is conducting. So in the end, we have to burn it off. Don't worry, it is very reactive. Any questions?

I guess you are asking what is next after FinFET? FinFET will take us to 10nm for sure. So you are asking what materials will be used for 7nm? I think that would be a good discussion.

<script src="//platform.linkedin.com/in.js" type="text/javascript">
lang: en_US
</script>
<script type="IN/Share" data-counter="right"></script>

 

[Li Yisuo]

7 is an odd number. 8 or 6 sounds much better from my superstitious view. Well RC turns to be critical, and air gap might be unavoidable.

what materials will be used for 7nm?

Are we ready for III-V valley or Ge Valley? I guess no, even though contact size scales to 40 now, simply too big. So mobility advancing probability <30%

what is next after FinFET?

Just two flashes @16/14 and 10, we would have to renovate our room again? What a luxury. To cut cost/leakage, solid-source sub-fin doping or even FINOI are cool, but not ideal. So keep increasing gate control to the limit, and you probably find my company. Bragging is a sin.
On the other hand, design cost mounts up too quickly. With current innovation constraints, it will take time for the market to figure out where the direction truly is.