Stories
Slash Boxes
Comments

SoylentNews is people

posted by CoolHand on Wednesday November 11 2015, @01:05AM   Printer-friendly
from the all-nanotubes-all-the-time dept.

The interplay of size and time may make carbon nanotubes the answer to the computer industry's prayers as it grapples with pressure to make silicon chips ever-smaller. Or the same factors may turn CNTs into a technological dead end.

Size refers to the dimensions of carbon nanotubes (CNTs) vs. the shrinking geometry of the components on today's silicon chips. A CNT is basically a tube whose wall is 1 carbon atom thick. The tube itself is 1 nanometer (nm, or one billionths of a meter, or one-thousandths of a micron) in diameter, although it can be tens of microns long. Although made of carbon, single-wall CNTs are excellent conductors thanks to quantum conductance, which allows electrons to propagate along the length of the tubes.

Time refers to the progression of Moore's Law, an observation by Intel co-founder Gordon Moore that the number of components on a chip can be expected to double every two years, without an increase in price. According to that, about more eight years from now silicon technology, which has reached 14nm geometry, will reach the atomic level. At that time, presumably the industry will no longer be able to uphold Moore's Law by making silicon components continually smaller.

Will CNTs, with their 1nm geometry, be ready by then?


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 3, Interesting) by Covalent on Wednesday November 11 2015, @01:34AM

    by Covalent (43) on Wednesday November 11 2015, @01:34AM (#261529) Journal

    Maybe nanotubes get us to 1nm from 14. But that's the end of the road. Since these are 14 times smaller, this would give us at most about 4 more Moore's Law doublings. Maybe 16 in 2D. That's still only 24-32 years max.

    The real frontier must be the third dimension. Even with existing silicon tech, moving in the z axis allows for incredible potential gains. No one has figured that out entirely yet, but we need look no further than the meat marvel in between our ears for both inspiration and justification. A computer cube (as opposed to a chip) could keep Moore's Law alive for centuries.

    --
    You can't rationally argue somebody out of a position they didn't rationally get into.
    Starting Score:    1  point
    Moderation   +1  
       Interesting=1, Total=1
    Extra 'Interesting' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   3  
  • (Score: 3, Interesting) by frojack on Wednesday November 11 2015, @02:44AM

    by frojack (1554) on Wednesday November 11 2015, @02:44AM (#261545) Journal

    But that's the end of the road.

    Famous Last Words.

    But anyway, I wonder how fragile something one atom thick might be.

    A tube whose wall is 1 carbon atom thick. The tube itself is 1 nanometer in diameter, although it can be tens of microns long.

    You would think one good knock would destroy them.

    With regard to the third dimension we've been doing that [wikipedia.org] for some time. Its kind of a mess, and vertical complexity is no where near what horizontal complexity is.

    If we can arrange nano-tubes in ANY orientation as we lay them down, they could well be that thing that allows chip layout in three dimensions through multiple layers.

    --
    No, you are mistaken. I've always had this sig.
    • (Score: 2) by Immerman on Wednesday November 11 2015, @06:40PM

      by Immerman (3985) on Wednesday November 11 2015, @06:40PM (#261855)

      >You would think one good knock would destroy them.

      Not really - they're moderately flexible, and held together by the strongest atomic bonds known to man. (graphene has an ultimate tensile strength of 130,000MPa, compared to 2,000-2,600MPa for steel.)

      Plus, not much delivers a focused knock at that scale. Hit it crosswise with the edge of a sharp scalpel and you'll probably break it though.

  • (Score: 2) by MichaelDavidCrawford on Wednesday November 11 2015, @03:45AM

    by MichaelDavidCrawford (2339) Subscriber Badge <mdcrawford@gmail.com> on Wednesday November 11 2015, @03:45AM (#261577) Homepage Journal

    We can go in the third dimension when we get rid of the thermal problems. 2-D works well because you can cool the whole surface of the chip, in 3-D the heat has to conduct through the body of it.

    It's not carbon nanotubes that will take care of that problem but high-temperature superconductors.

    --
    Yes I Have No Bananas. [gofundme.com]
    • (Score: 2) by AnonymousCowardNoMore on Wednesday November 11 2015, @03:08PM

      by AnonymousCowardNoMore (5416) on Wednesday November 11 2015, @03:08PM (#261745)

      Resistance is still needed and in any case the computer must at least dissipate the amount of heat mandated by quantum mechanics to flip a certain number of bits. (Very) high-temperature superconductors would certainly help if they could be used in chips.

      I can't help but wonder how feasible it is to make the substrate of diamond. Extremely high [wikimedia.org] thermal conductivity combined with low electrical conductivity seems ideal for the job if it can be done economically. Some googling indicates that there is research interest in the matter.

    • (Score: 2) by bob_super on Wednesday November 11 2015, @04:44PM

      by bob_super (1357) on Wednesday November 11 2015, @04:44PM (#261813)

      Actually, I believe IBM has prototypes of micro-channels through the wafer for liquid cooling. If you can cool the chips horizontally, then you can stack them with TSVs and get significant "3D" (quotes because each layer is still flat) benefits.

    • (Score: 2) by Immerman on Wednesday November 11 2015, @06:52PM

      by Immerman (3985) on Wednesday November 11 2015, @06:52PM (#261860)

      Well, high-temperature superconductors or thermally conductive semiconductors. That's one of the reason diamond-based semiconductors hold so much potential - diamond is an extremely good thermal conductor (about 3x better than copper), while also being an extremely good electrical insulator. Plus we know how to perform both N- and P-doping on it to produce efficient semiconductors. A rare combination of properties.

      At this point we know how to mass-produce it in flawless slabs (even the best natural diamonds contain far too many flaws to consider as a substrate, even if they weren't tiny). It's just a matter of waiting for the current "seeds" to be grown to sufficient size that they can start slicing off slabs suitable for manufacturing, or developing new, faster methods of producing flawless diamond. I think it's carbon-vapor deposition being used by the current forerunners, (the makers of "cultured" diamonds) and if I recall correctly their slab size only increases by a few percent per year, with each new layer being just slightly larger than the last.

  • (Score: 0) by Anonymous Coward on Wednesday November 11 2015, @03:47AM

    by Anonymous Coward on Wednesday November 11 2015, @03:47AM (#261578)

    The trouble with 3D circuits is heat dissipation, 2D chips can leak heat out the top, but if you build more circuit that way the heat has farther to go before it gets to the heat sink.

    That means you have to run the IC slower so it doesn't get as hot.

  • (Score: 2) by mhajicek on Wednesday November 11 2015, @05:22AM

    by mhajicek (51) on Wednesday November 11 2015, @05:22AM (#261614)

    What I'm wondering is when someone will find a way to cram multiple functional units per atom. Remember the atom (despite it's name) is not the fundamental building block. Perhaps electron shells can be accessed independently, or overlapping waveforms or somesuch. As history keeps showing, just 'cause we don't know how doesn't mean someone won't figure it out.

    --
    The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
    • (Score: 0) by Anonymous Coward on Wednesday November 11 2015, @10:59AM

      by Anonymous Coward on Wednesday November 11 2015, @10:59AM (#261685)

      The problem with using energy shells is that excited states decay very quickly. There might be a way to do interesting things with the spins, though.

      • (Score: 2) by bd on Wednesday November 11 2015, @04:53PM

        by bd (2773) on Wednesday November 11 2015, @04:53PM (#261817)

        Spin polarisation dephases even more quickly if you are significantly above 0 K. That is pretty much the fundamental problem of spintronics.