[HN Gopher] Researchers create first functional semiconductor ma...
___________________________________________________________________
Researchers create first functional semiconductor made from
graphene
Author : wglb
Score : 98 points
Date : 2024-01-08 14:21 UTC (1 days ago)
(HTM) web link (techxplore.com)
(TXT) w3m dump (techxplore.com)
| mdaniel wrote:
| original source discussion:
| https://news.ycombinator.com/item?id=38878780
| ptek wrote:
| With 5nm silicon process, I wonder where this is going to start
| with retooling and building new machines? I'm guessing they will
| start with 8bit RISC cpus with maybe the first 128 bytes as a
| data cache which could also be used as extra registers?
|
| It will be great having a lot of expired patents and 50+ years of
| silicon work to build on, Machine Learning tools and free
| operating systems (NetBSD, Linux)
|
| I hope they do it all again, I want to relive the C64, Amiga and
| 680x0 era again.
| thijson wrote:
| This will start with small devices used in power electronics
| likely. There are faster materials than silicon, but they still
| aren't used for highly integrated digital logic due to cost.
| logicchains wrote:
| If starting from scratch with new tooling, I wonder if they'll
| consider ternary, which is theoretically more efficient than
| binary (as it's closer to the optimal encoding: e-ary).
| BoiledCabbage wrote:
| I believe it, but do you have any links to somewhere
| discussing why e-ary is the optimal encoding?
| f_devd wrote:
| It's likely a reference to the radix economy [0], however
| it's actually very minor for the cases where it works. And
| in "the best way to count"[1] there compelling arguments as
| to why the formulation used on Wikipedia is not necessarily
| correct for arbitrary precision numbers as well as being
| bad for human notation compared to binary.
|
| [0]: https://en.m.wikipedia.org/wiki/Radix_economy[1]:
| https://www.youtube.com/watch?v=rDDaEVcwIJM
| ahf8Aithaex7Nai wrote:
| It's even closer to pi-ary
| NegativeK wrote:
| I can't believe that they'd do that, since it'd mean
| rewriting huge swaths of existing software.
| bilsbie wrote:
| How this made? Still with lithography?
| crakenzak wrote:
| Someone please make me excited about this, what can we
| potentially see from this? How soon until potential
| commercialization?
|
| Is this a technology that will carry us over a currently
| uncrossable river?
| aabhay wrote:
| Graphene has been a long standing aspiration in semiconductor
| materials science:
|
| - better current load capability - better heat dissipation
| characteristics
| sim7c00 wrote:
| sorry for my being 8 year old, but, does this better current
| capacity mean it can go faster? and the less heat problems
| mean it can do so without being cooled riduculously? (i.e.
| faster consumer products??)
| droopyEyelids wrote:
| Computing with graphene is still sci-fi territory, making a
| semiconductor only allows you to make transistors, not to
| necessarily make billions of transistors at nano-scales.
|
| On the larger physical scale, transistors are used for
| stuff like mosfets and switching power supplies.
|
| You might have noticed the new generation of Gallium
| Nitride (another semiconductor) USB chargers. They improve
| over silicon in every important dimension, I think, with a
| higher breakdown voltage, lower on resistance, higher
| electron mobility (which is what limits the speed it can
| cycle).
|
| The gallium nitride chargers (when well engineered) waste
| less power, meaning they don't get as hot, and can be built
| much smaller for a given output.
|
| I am not an engineer or a materials scientist, but I think
| graphene is better in thermal conductivity and electron
| mobility than GaN, but a semiconductor would probably have
| a smaller bandgap, because graphene naturally doesn't have
| much of one at all, meaning it'd have a smaller operating
| range in temperature and voltage.
| throwaway4aday wrote:
| Probably the most exciting thing about this is that it shows
| the hard work of labs to discover the tacit knowledge necessary
| to produce single layer graphene is bearing fruit. Developing
| the real world industrial processes for putting theory into
| practice is always the hardest part of creating new technology
| especially when doing so on atomic scales. It seems slow and
| underwhelming compared to whizbang announcements about wonder
| materials but this is the type of gradual progress that the
| future is based on.
| drekipus wrote:
| > Someone please make me excited about this
|
| Why?
| rustcleaner wrote:
| Incept running GrapheneOS on graphene dies.
| vanderZwan wrote:
| > _" To me, this is like a Wright brothers moment," de Heer said.
| "They built a plane that could fly 300 feet through the air. But
| the skeptics asked why the world would need flight when it
| already had fast trains and boats. But they persisted, and it was
| the beginning of a technology that can take people across
| oceans."_
|
| Tangent: I always find comparisons to the Wright brothers kind of
| awkward, because everyone seems to overlook their subsequent
| patent war and how much damage that did to the aviation industry
| in general (tangent: similarly Edison really _really_ stifled
| early innovation in lightbulb technology with patent trolling.
| Philips became as big as it is because there was a decade in the
| 19th century where the Netherlands didn 't recognize patent laws,
| so was free to steal and combine every innovation out there. They
| also employed highly motivated former rivals of Edison whom the
| latter had put out of business)
|
| Anyway, with that side-track out of the way: I don't want to
| sound to negative, I do hope this is the breakthrough it's hyped
| up to be!
|
| [0] https://en.wikipedia.org/wiki/Wright_brothers_patent_war
| Xeamek wrote:
| For a second I thought that this conductor somehow supports
| functional programming instead
___________________________________________________________________
(page generated 2024-01-09 23:00 UTC)