You are cherry picking. In 2008/9 they were acquired by Lockheed Martin. They have commercial applications out *ALREADY*, as noted in the article. The "new" company is the commercial front. Fujitsu has been working on taking it to mass production next year. Few others will follow
-
-
If you mean the containment of CNTs in a cleanroom to prevent contamination, skywater has already shown this to be possible
-
They said at HC they already manufactured 10s of 1000s of NRAM dies. I'm sure they understand the manufacturing problems involved.
-
One could say the same thing of Intel’s 10nm process. The road to high volume isn’t easy, especially when new materials or structures are involved.
-
Very true.
End of conversation
New conversation -
-
-
So the idea is the entire layer is cnt? Not just cnt in cells? How do they planarize? How do you contact through cnt? How do they do top contacts? Like do you etch stop in the cnt and if so how does the oxidized cnt effect things? Just potential integration challenges
-
Per this it looks like they just fill the cells with nanotubes but the rest of the layer is dielectric. How do they remove the excess CNT from the top of the dielectric outside the cell. It can't be a CMP. I'm sure they figured it out but how? https://fuse.wikichip.org/wp-content/uploads/2018/09/nram-scaling.png …
-
Why not CMP?
-
Carbon nanotubes are like barely cooked spaghetti. CMP is a rotary sander. What happens if you rotary sand barely cooked spaghetti? I'd guess you'd get a lot of tearing as you rip out tubes causing a lot of voids and pretty low yields. Organic stuff is pretty soft for CMP..
-
It's way less dense but they use off-the-shelf standard coating tools you can find in any fab (looks like CMP to me). It boils down to fine process controls and experience working with the mixture.
-
Where do you see it's CMP? I mean if you spincoater any polymer even without high aspect additives over trenches you will get material outside the trenches.
-
I attended a presentation of theirs a while back (before hot chips) which had a few additional pictures of the flow. IIRC it looked like CMP. Could be wrong though.
-
Hmm if you happened to remember when I'd love to see their flow. As I said I'm sure they figure it out but we do a lot of research into polymer thin films so it would be nice to see their approach.
- 1 more reply
New conversation -
-
-
Apologies if this is a dumb question: Does the nanotube slurry actually physically move in operation? If so, does it not create mechanical stress on the electrodes over time?
-
Yes it physically moves in a void that is formed. But only a certain amount of CNTs that were deposited at the bottom of the cell participate in the 'storage'. The rest are fixed to the parameter of the cell which prevent stresses and other materials from permeating into the cell
-
Thanks for explaining
-
If this is real, then it potentially renders Samsung, SK and Micron's products obsolete. Why is Nantero content to be a design house??
-
Try pitching a startup that needs 10 billion for a new fab :)
-
I am ~80% sure their Lockheed Martin acquisition has locked them into this model due to some inter-licensing restriction. But who knows.
-
Isn’t it true that they can pursue a fabless model? At least initially? Besides the TAM is so huge, even raising $10B might not be impossible.
-
Or, they can pick one of the DRAM guys for a JV. Similar to what Sandisk did with Toshiba. If it is as good as it looks, they have massive bargaining power.
- 2 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.