Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
User Journal Technology

Carbon Nanotube-Based NVRAM In 2-3 Years? 66

Posted by ScuttleMonkey
from the nano-ram dept.
According to NanoWerk, UC Riverside researchers have come up with a memory device based on telescoping multi-walled carbon nanotubes. According to one of the researchers, 'This finding leads to a promising potential to build ultrafast high-density nonvolatile memory, up to 100 gigahertz or into the terahertz range" and a prototype could be demonstrated "in the next two to three years.' Similar devices from UCLA and Caltech based on bistable rotaxanes are farther along in being integrated into actual memory circuits, but tend to break after a fairly small number of position changes. Carbon nanotubes may promise more durable switches.
This discussion has been archived. No new comments can be posted.

Carbon Nanotube-Based NVRAM In 2-3 Years?

Comments Filter:
  • by Stile 65 (722451) on Wednesday February 14, 2007 @01:30PM (#18014450) Homepage Journal
  • by StressGuy (472374) on Wednesday February 14, 2007 @01:31PM (#18014460)
    "Backend Architecture, Nano-RAM, type A" it would be called BA-Nano-RAM-A? ....it's ok....I'll just go now....
  • by haluness (219661) on Wednesday February 14, 2007 @01:36PM (#18014530)
    A brief scan of the paper seemed to indicate that their results were based on simulations. Do they have some working model that justifies it coming out in 2 or 3 years?

    Or did I read the paper to fast (hey, at least I *did* RTFA)?
  • who knew (Score:5, Funny)

    by President_Camacho (1063384) on Wednesday February 14, 2007 @01:36PM (#18014534) Homepage
    According to NanoWerk, UC Riverside researchers have come up with a memory device based on telescoping multi-walled carbon nanotubes.

    Who would have guessed that, in the future, your computer would be a series of tubes?
    • Who would have guessed that, in the future, your computer would be a series of tubes?

      Yeah, but I'd rather it were a big truck--something I can just dump something on.
    • by sokoban (142301)
      ultrafast high-density nonvolatile memory
      Apparently it will be a series of tubes which you will be able to just dump something on, because the tubes will be able to store an enormous amount of material, enormous amount of material.
    • by ch-chuck (9622)
      That's what ENIAC was.
      • ENIAC's tubes were in parallel.
        • Re: (Score:2, Informative)

          by ch-chuck (9622)
          I think it did a lot of work as a serial processor. From this description [upenn.edu]:

          "The ENIAC was controlled through a train of electronic pulses."

          --and--

          "because the various units of the ENIAC could operate simultaneously, the ENIAC could perform calculations in parallel. (BUT!) ENIAC programmers tended to avoid this use because the impressive but limited reliability of the ENIAC favored the use of as few units as possible for a given application."
    • I, for one, welcome our high-density, nonvolatile, terahertz memory nano-tubular overlords!
    • Well, at least we know that it - like the Internet[s]- is not a dump truck.
  • Nano Abacus? (Score:4, Insightful)

    by Radon360 (951529) on Wednesday February 14, 2007 @01:40PM (#18014588)

    It looks to me like they've essentially created what could be compared to a nano-abacus. I wonder how immune this system would be to physical movement (i.e. jarring). In a similar vein, I would imagine that it would be just as static sensitive as most other memory devices even though.

    Did I miss something, though? How is the position of the telescoping tube read? Applying a current to it would change the position, would it not?

    • Re: (Score:2, Insightful)

      by unc0nn3ct3d (952682)
      I imagine this would be just as immune to physical jarring as say blood cells in your blood vessels would be wouldn't it? Or as resiliant as the atomic bond in elements are.. At this scale the physical movements that we create as humans wouldn't be felt, similar as the molecules in your hand don't feel it when you wave at someone, but the hand as a whole feels it.. or at least that is what I would think
    • by owlstead (636356)
      Ehm, just a thought: first you put a possitive V on one side, making it go to the right (picture b). After that you put a (higher) possitive voltage to the outer tube (note that it is also connected, initially to the ground). The left simply acts as ground, not atracting the inner tube. Could this work? This would also mean that the tube would stay attracted to the right, solving the first problem you came up with.

    • by IdahoEv (195056) on Wednesday February 14, 2007 @03:56PM (#18016258) Homepage
      At the nano scale, momentum of objects is near zero and friction forces, van der waals, and the like dominate entirely. Macro-scale motion, and even intense vibration, simply won't move things around relative to each other.

       
  • A new take on an old concept? Using nanotubes?! Who'da thunk it!
  • Carbon's pretty good. I assume you could use diamond in it's purest form to make the basis of a processor after doping it properly to make it a transistor. Thermal tolerances for these would be excellent. But I think I'll wait for Ovonic Unified Memory. It's already technology in use today, just done a bit differently for the applications I'm waiting for it to be used for.
    • by cyfer2000 (548592)
      Diamond as semiconductor has been studied for some time. There are some progress [aist.go.jp] recently. But it is not easy to make big single crystal diamond yet, not to mention single crystal diamond wafer. Just imagine how difficult it could be to polish the diamond wafer without introducing many defects, and how hard could it be to characterize how many defects you get in the wafer.
    • You oust an existing technology, you need more than a high coolness factor. NAND flash costs less than 2c per MByte and falling. There are plenty factories set up to produce it, it is fast enough and low power enough for most mobile applications.

      In 2-3 years there might be a nanotube demo, but that's a long way from being something that you can mass produce for significantly lower cost than NAND.

    • Re: (Score:2, Informative)

      by chrisb33 (964639)
      I wouldn't condemn Nantero to vaporware status just yet - it seems that they've been making progress. Here's a list of their press releases [nantero.com] - notice that they successfully fabricated a switch in April and have made their processes compatible with current CMOS fab lines.
  • by mark-t (151149) <markt@@@lynx...bc...ca> on Wednesday February 14, 2007 @02:00PM (#18014816) Journal
    I've just seen that figure too many times now... 2 years is still a short enough time that it might seem feasable but still long enough away that by the time it has gone by, most everybody will have forgotten about it and moved on to something else.

    It'd be really neat if this turns out to be genuine, but I'm not holding my breath. Been disappointed too many times already.

    • This technology http://www.nantero.com/mission.html [nantero.com] is suppose to be ready some time this year. It use cnt as lever in a small relay. They call their memory nram and it is suppose to be very fast access and non-volatile. There is a video at their site explaining how its memory works.
    • by Kjella (173770)
      I've just seen that figure too many times now... 2 years is still a short enough time that it might seem feasable but still long enough away that by the time it has gone by, most everybody will have forgotten about it and moved on to something else.

      2-3 years to early prototype
      5 years to well working prototype
      7-8 years to get it to mass production
      10+ years to consumer markets ...even if this stuff is true, it's far off in the future. But, I guess we can hope...
  • Prescient SF FTW (Score:4, Insightful)

    by JesseL (107722) on Wednesday February 14, 2007 @02:01PM (#18014836) Homepage Journal
    Sounds vaguely similar to the nano-scale rod-logic of Neal Stephenson's stories.
  • by LM741N (258038) on Wednesday February 14, 2007 @02:22PM (#18015080)
    Memory and/or processors running at 100Ghz sounds great, but how is such a chip going to be connected to the outside world of peripherals? Beams of light? Waveguides? Or will everything have to be contained on one chip?
    • by JesseL (107722)
      I/O should never happen at the same speed as the processor. It has to be this way in order for the processor to actually process anything fast enough to keep up with whatever it's connected to.

      So the answer to your question is: The same way it always has, through things like buffers and UARTS.
    • entwined pairs [arxiv.org] of course! duh!
    • Not to mention that whatever you interface it to has to be incredibly broadband. To achieve the edge-times on those 100GHz square pulses, you'd need a hell a lot of harmonics.
  • I dont doubt it might be cool technology, but 2-3 years isnt too realistic...
  • Thank you for the link to the Wikipedia article on rotaxanes.

    The name, rotaxane, is derived from the Latin for wheel (rota) and axle (axis).

    Ah, so I must have been incorrect in initially parsing it as being pronounced with a silent- or H-like X. Am I the only one to read it that way?

    It also took me a moment to parse "bistable" as "bi-stable" instead of "bis-table". (Don't tell me it should be "bist-able".)
    • Re: (Score:2, Funny)

      by Rethcir (680121)
      Rotaxaaaaane! You don't have to put on the red light!
    • This is a growing problem. Far too many terms are being introduced with uncertain pronunciation, and the initiators really should provide phonetics. (It's interesting that even the NATO/ITU alphabet needs phonetic explanation because it uses words like "Charlie" and "Whiskey")

      Apart from Linux (Linnux? Leenux? Lie-nux?) there are the old saws of schoolteachers - words like periodate, unionised, benzoyl - and even simple looking words like "kilometre" - kilo-metre or kilom-eter?

  • Rotaxane?!? (Score:1, Funny)

    by Anonymous Coward
    Wasn't that the song The Police played to open the Grammys? Damn, those guys were years ahead of their time!
  • This model is double walled nanotube only, any body know any method to grow specifically double walled carbon nanotubes?
  • What part of 'This finding leads to a promising potential to build ultrafast high-density nonvolatile memory, up to 100 gigahertz or into the terahertz range" and a prototype could be demonstrated "in the next two to three years".' reminds me of the Monty Python comment in a spoof ad that "up to" includes zero?

    When a paper is as full of weasel words as this one, reach for your Dilbert collection.

  • Of course! (Score:1, Redundant)

    by NerveGas (168686)

        Just like that holographic storage that was going to replace hard drives "in 2-3 years"... almost a decade ago.

        That's not to say that this *won't* happen, just that it's yet another "We're going to change the world in a few years!" idea, which should really be a "We'll wake you up if anything ever becomes of this." sort of message.
  • Did noone else notice that TFA referred to hard disks as "sequential access"? I thought that was tape...
  • Wow. It's been a long time since I've read an abstract with that little information in it. What a vapid piece of fucking trash. I'm sure that the next paper out of that research group will be "Telescoping carbon nanotube (CNT) space elevators in 3-5 years".

    No data. No numbers. No quantification, no discussion of advantages or difficulties with the technology. No discussion of the fact that it is quite frigging difficult to get relatively defect-free nanotubes in any sort of practical volume. Somethin
  • It's one thing to make one of them. Economically making arrays containing enough working components to provide commercially viable products is another thing altogether.

Those who can, do; those who can't, write. Those who can't write work for the Bell Labs Record.

Working...