Single-Atom Layer of Tin May Be a New Wonder Conductor 126
At Kurzweil AI, an article proclaims that the next wonder material for computer chips may be an unexpectedly common one:
"Move over, graphene. 'Stanene' — a single layer of tin atoms — could be the world’s first material to conduct electricity with 100 percent efficiency at the temperatures that computer chips operate, according to a team of theoretical physicists led by researchers from the U.S. Department of Energy’s (DOE) SLAC National Accelerator Laboratory and Stanford University." (Original paper is available here, but paywalled.)
100 percent efficiency? (Score:4, Insightful)
Seems unlikely. Something about complimentary midday meals...
Re:really? (Score:2, Insightful)
Somehow i also doubt the claim "replacement for silicon", silicon is not used in IC-s for its conductivity properties, au contraire, its used because it can be doped to become a N or P semiconductor. Interconnects in IC-s are made with plain old aluminum and copper and interconnects are really not the point where power is consumed. Lion share of power in FET based logic is used charging and discharging transistor gates, that is the losses are capacitive, you can reduce the losses by making smaller transistors but you really cant affect it by material selection.
Re:really? (Score:5, Insightful)
It is right and proper to have doubts about new announcements like this. That is the basis of science - the idea of "replicate, then trust, but verify" at the core of scientific approaches. If this turns out to be either an error, a late April-fools joke, a scam, a one-off result that cannot be replicated, or a valid result within a small range of constraints, then it will be labelled as such.
However, if subsequent independent experiments show a robust and consistent process that can be replicated easily, then I for one will welcome our new (1 atom-thick) tinfoil hat-wearing overlords...
Re:really? (Score:2, Insightful)
Usually science in this sort of parers is valid. Just that journalists "translate" it to load of crap and attach applications to the science that were never part of the original paper.
It's all simulations! (Score:5, Insightful)
At least as far as I can tell without access to the paywalled concept.
Important questions would be:
What is the maximum current that can be transported through strips of various widths?
How sensitive to defects is the process?
Tin is going to be a major problem for much semiconductor processing - as it means you basically now can't solder the chip, or do any even 'low' temperature processing after it's deposited - it has to be the last layer.
Re:really? (Score:3, Insightful)
"Trust" and "verify" are contradictory. It's fine that you want to verify, but don't pretend that you are trusting while you actively violate the concept of trust.
Re:really? (Score:5, Insightful)
Bzzzt! wrong. If I trust you, I will verify your work. If I don't trust you, I won't even bother to look at it.
Trust is about honesty, not about infallibility.
Re:really? (Score:4, Insightful)
If you don't really trust anything you haven't personally verified yourself, you can't get very far in real life. Have you verified every line of code that runs on your computer or do you trust it enough to run it anyway? What about the compiler? Do you ever use results from your coworkers without digging through every calculation down to every assumption and verify them? Trust is a measure of confidence in your word, which is weighed against how important it is. If you have no confidence in it, you don't want to waste any time on it (assume false, test false). If you have a bit of confidence that it's might be worthwhile you verify it. (assume false, test true). If you think it's probably right, but you want to verify it that's stronger (assume true, test true). And if you trust it implicitly that's of course trust (assume true, test false).
Re:really? (Score:4, Insightful)
In day-to-day life you have to trust almost everything to get anything done, but the GP was talking in particular about trusting new scientific results. Even then if several independent scientists have verified it I can trust it without verifying it myself, but trusting an initial scientific finding that no one else has verified is just foolish, no matter how smart and established the original scientist is. There are different levels and meanings to the word "trust", and as GP pointed out there has to be a certain level of trust for anyone to even bother to try verifying or debunking a result (because there are far too many crackpots out there to deal with them all).