Is this a contradiction...its both a conductor and an insulator? And if it's both, why not eliminate the copper completely? The linked article doesn't explain this... or maybe I'm just too obtuse to understand it.
Scientists at Stanford have found a new use for graphene that will significantly increase the speed of standard computer processors. Most chips use copper interconnects to route signals around their semiconductor circuits, and these are typically insulated with a coating of tantalum nitride. The Stanford team instead used an …
" That was my thought too, if it's so much better than the copper just ditch the copper."
I was going to write the same thing!
Is it maybe that the graphene is an insulator to copper, or to put it another way: you could use graphene for the interconnects, but then you'd need something (say copper) as the insulator! (But even then, why not some non-metal insulator?)
Baaaah. I'm confused!
not a contradiction at all.. electrons will flow much easier within the plane of the graphene sheet than jump from the sheet to the nearest conductor. It wouldn't surprise me if the resistance within the plane of the graphene is actually lower than in the copper main conductor locally..
It's like the old trick with the tablecloth.. If you pull it away fast enough things on it will stay on the table.. mostly..
The same "trick" is often used in nature in a host of enzymes, actually.
Thanks for the explanation. So basically, the electrons don't penetrate the grapheme but ride along on the surface next to copper? If voltage were applied to the exterior that were they would ride?
Given the speeds most processors, etc. run at, I would have expected to see some RF doing the surface ride also on the copper.
Contradictory or not.. this is fascinating. I hope they can pull this one off and use it commercially.
This post has been deleted by its author
It would work that way if you consider the graphene not as a sheet of carbon, but two layers of hydrogen stabilised by a lattice of carbon + electron soup ( extremely simplified, of course).
To migrate to the silicon any copper atom would have to react with the H of the graphene first.. and there it stops, really. Given the purity of the environment, only direct synthesis would be possible, and while the gods of Quantum could pull a fast one and provide the energy, the local temperature would ensure any CuH formed would revert near-immedeately. ( reverse reaction "happens readily" at 20 K, let alone 20 C...)
Then there is the issue of inserting into/breaking the benzene-ring lattice itself. And the Si-facing layer of H in the sheet ( with even more ...impressive... energies in play..). Let's just say the odds are stacked against it.
"It would work that way if you consider the graphene not as a sheet of carbon, but two layers of hydrogen stabilised by a lattice of carbon + electron soup ( extremely simplified, of course)."
Not sure the point you are making - It's known that a monolayer of graphene is an impermeable barrier under normal conditions ( I have a feeling that protons can tunnel through ) So as I stated copper is not going to be able to diffuse into the silicon
"f you consider the graphene not as a sheet of carbon, but two layers of hydrogen stabilised by a lattice of carbon + electron soup ( extremely simplified, of course)."
"To migrate to the silicon any copper atom would have to react with the H of the graphene first.."
Sorry I should have noticed this sooner but I've been busy today. I'm afraid you are under a misapprehension - graphene doesn't have any hydrogen attached to it - it's pure carbon sp2 hybridized. so planar like benzene but with all bonds carbon-carbon
I could be mistaken, but I was under the impression that graphite (and by extension, graphene) was an electrical conductor but a heat insulator.
Whereas diamond was a heat conductor but an electrical insulator.
Of course, I was "told" that some years ago and have probably misunderstood and misremembered it.
... Why not eliminate the copper completely?
Perhaps because graphene is one atom thick and has no structural strength? Ergo it needs to be layered on something that does. Why copper? Presumably for practical reasons, for instance because copper is the known quantity and allows an apples-to-apples comparison between graphene and traditional coatings, because common sense says don't change more than one variable at a time if you don't have to, and so on.
Why assume this is anything other than the usual incremental evolutionary step forward from the current state of the art?
@chemist : I stand corrected. Strongest material known to man. Yet it seemed so puny on the Sellotape! Icon is for self. Need bigger ship ^^^, etc.
Moving on, I'm sure you can improve on my answer, if you care to. If the structural strength of graphene is currently realisable (no pun intended), why not ditch the copper entirely and go with a graphene conductor? I don't see any other commentards try to answer that one.
I suspect my answer was otherwise substantially correct in terms of practicalities, but full disclosure, I'm not really Sherlock, just another random commentard idly speculating
You can buy adhesive "Pyrolytic Graphite" sheets for large sums of money. It's a flexible, lightweight, graphene-like material that conducts heat incredibly well. Cellphones sometimes have a strip of it down the back so the CPU and GPU chips don't burn your hand. Ultra-compact switching power supplies use it to cool components that are spaced closer together than normally possible.
This post has been deleted by its author
Dunno. Be more than a single atom thick? It's interesting to know all the things it can do, but moreso to know what it does well and in fact better than other materials. In the case of this article, it seems that we have a winner. Next comes scalability and cost on the checklist to determine if it is worthwhile to implement this.
Pyrolytic graphite isn't graphene though it does contain "bits of" graphene as a product of its production process (in the same way that a cabbage contain lignins). The whole point of graphene is that it is a continuous sheet - and the problem with that is there are no interconnects (ie lattice defects) between the sheets to reduce the strength. Pyrolytic graphite is an impure source of graphene which allows lots of intra-lattice bonding - less strong, less electrically conductive but more usable and much, much cheaper than pure graphene.
Pure graphene is actually really useful for ... err ... and ... ummm ... other that graphene is useful for and obviously other things too. And it's a brilliant material because the His Royal Highness Cameron is investing loads into it and I totally trust his judgement with tax-payers cash.
"And it's a brilliant material because the His Royal Highness Cameron is investing loads into it and I totally trust his judgement with tax-payers cash."
In the interests of calibrating my sarcasm detector...
Would that be the same David Cameron who decided that drug policies should not be guided by evidence gathered from scientific studies because he personally disagreed with the conclusions ?
Biting the hand that feeds IT © 1998–2020