Forgot your password?
typodupeerror
Data Storage Upgrades Technology

Researchers Create Graphite Memory 10 Atoms Thick 135

Posted by timothy
from the comes-with-convenient-pink-rectangular-prism dept.
CWmike writes "Researchers at Rice University have demonstrated a new data storage medium made out of a layer of graphite only 10 atoms thick. The technology could potentially provide many times the capacity of current flash memory and withstand temperatures of 200 degrees Celsius and radiation that would make solid-state disk memory disintegrate. 'Though we grow it from the vapor phase, this material [graphene] is just like graphite in a pencil. You slide these right off the end of your pencil onto paper. If you were to place Scotch tape over it and pull up, you can sometimes pull up as small as one sheet of graphene. It is a little under 1 nanometer thick,' Professor James Tour said."
This discussion has been archived. No new comments can be posted.

Researchers Create Graphite Memory 10 Atoms Thick

Comments Filter:
  • by Warll (1211492) on Thursday December 18, 2008 @07:56PM (#26167381) Homepage
    As an optimist myself I would have said that it was 10 atoms thin!
    • by aztracker1 (702135) on Thursday December 18, 2008 @10:27PM (#26168499) Homepage
      I was told thicker is better... ;)
      • Re: (Score:2, Funny)

        by NerdyLove (1133693)
        More to love :-)
      • by Mozk (844858) on Friday December 19, 2008 @12:01AM (#26169143)

        When cornered into a room by ninjas with nothing separating you from them but a door of wood, yes, thicker is better, but you will die regardless.

        • Re:Ninjas? (Score:5, Funny)

          by aywwts4 (610966) on Friday December 19, 2008 @06:47AM (#26170963)

          When cornered into a room by ninjas with nothing separating you from them but a door of wood, yes, thicker is better, but you will die regardless.

          I think you are confusing ninjas with zombies, zombies have thick wood door shredding powers while a ninja is already in the room with you.

          • by Gilmoure (18428)

            So does John Wayne.

            In one incident [imdb.com], Bond bet Wayne that they could stand on opposite sides of a newspaper and Wayne wouldn't be able to hit him. Bond set a sheet of newspaper down in a doorway, Wayne stood on one end, and Bond slammed the door in his face, shouting "Try and hit me now!" Wayne responded by sending his fist through the door, flooring Bond (and winning the bet).

        • by Myrddin Wyllt (1188671) on Friday December 19, 2008 @08:48AM (#26171505)

          I'm not sure thicker is better. I remember hearing that churches in northern England replaced their super-thick oak doors with thinner planks riveted together in a cross-ply design, as this provided better protection against the axes of marauding Vikings.

          Of course, Ninjas are a different proposition, and five minutes googling gives me no citation for the monastic plywood theory, so perhaps direct experiment is the only way to settle this one - just make sure you have plenty of emergency Pirates on hand for back-up and it should be safe enough.

    • by dissy (172727)

      Researchers at Rice University have demonstrated a new data storage medium made out of a layer of graphite only 10 atoms thick.

      640k of atoms should be thin enough for anybody!

  • Finally.. (Score:3, Funny)

    by jmerlin (1010641) on Thursday December 18, 2008 @07:57PM (#26167387)
    I store data using just a pencil, paper, and some tape. I knew there was a way. Oh wait...
    • by onion2k (203094) *

      What's the tape for?

      • by Anonymous Coward on Thursday December 18, 2008 @08:28PM (#26167673)

        Compression.

        You fold the paper in half, and then tape the ends. Voila! Same information, half the size!

        • Re: (Score:3, Funny)

          by Todd Fisher (680265)
          Yeah but you can only do that 8 times. Pfft some technology!
          • Re:Finally.. (Score:5, Informative)

            by ecalkin (468811) on Thursday December 18, 2008 @10:41PM (#26168581)

            you should watch some mythbusters!

            i think they managed 12 or 13 folds.

            of course they started with a sheet of paper the size of a house and made the last fold with the help of heavy machinery!

            eric

            • I can do at least 25 folds.

              Think accordian...

            • Re:Finally.. (Score:5, Informative)

              by hvm2hvm (1208954) on Friday December 19, 2008 @05:36AM (#26170659) Homepage
              No, the huge thin sheet only got to 8 and they needed a forklift to fold it. 12 folds would take an extremely large (or very thin) sheet of paper. That's because folds make the paper exponentially thicker and smaller. So, for the same thickness for each new fold you need to make the paper 2 times exponentially larger. I'm too lazy to think whether it's something like x^2^2 or (x^2)^2 (or just x^2 since you fold it along width and height alternatively). Anyway it grows fast since an A4 sheet can be folded 7 times and a warehouse sized thinner sheet gets to 8.
              • Re: (Score:2, Informative)

                by KiwiCanuck (1075767)
                The force required to deflect/bend a beam is proportional to the cube of the thickness. Each time you fold the paper you make it twice as thick, and therefore 8 times stronger.
        • - You repeat this process infinite many times, thus solving the problem once and for all.
          - But, but...
          - ONCE AND FOR ALL!

          (Also, about 37 foldings of it would make the paper so high to reach the moon).
          • by smoker2 (750216)
            7 times - try it.
            • by Arterion (941661)

              I think he means that $width_of_paper ** 37 >= $distance_between_earth_and_moon.

              • Re: (Score:3, Informative)

                by volsung (378)
                Almost. That should be $width_of_paper * 2**37.
              • by setagllib (753300) on Thursday December 18, 2008 @11:01PM (#26168705)

                Every time you use an unspecified unit as the base in an exponential function, baby Newton cries.

              • by Daimanta (1140543)

                "$distance_between_earth_and_moon."

                European or African moon?

            • 7 times - try it.

              more than 7. [wikipedia.org]

              • "A special kind of $85 toilet paper"

                I'm glad she didn't use the regular $85 toilet paper. But outside of paper-folding experiments, who actually uses this? Does it really feel so much better?

                • by Gilmoure (18428)

                  But outside of paper-folding experiments, who actually uses this? Does it really feel so much better?

                  It's so good, it actually does my calc homework for me.

          • by vux984 (928602) on Thursday December 18, 2008 @09:37PM (#26168209)

            (Also, about 37 foldings of it would make the paper so high to reach the moon).

            No problem. Just bend the resulting column in half 37 times.

            ONCE AND FOR ALL!

  • Space Exploration (Score:5, Insightful)

    by Szentigrade (790685) on Thursday December 18, 2008 @07:58PM (#26167397)
    This could be a real boon to space exploration. Temperature extremes and radiation are two of the most common problems that must be dealt with when designing exploratory vehicles. This could simplify things greatly.
  • 10 Atoms thick? (Score:2, Interesting)

    by Anonymous Coward

    As per wikipedia,

    Diameter range: 62 pm (He) to 520 pm (Cs) (data page)

    Atom @ Wikipedia [wikipedia.org]

    It seems that the "thickness" of an atom varies. I've never understood why it is used as a unit of measure.

    • Re: (Score:2, Informative)

      by Anonymous Coward

      In this instance, it seems highly likely that they're referring to atoms of carbon as those are the atoms which compose the material involved.

      • Re: (Score:1, Informative)

        by Anonymous Coward

        In this instance, it seems highly likely that they're referring to atoms of carbon as those are the atoms which compose the material involved.

        Interesting, let me see.

        Carbon [wikipedia.org]

        They say an atom of Carbon is about 80 pm (picometers) in diameter. A picometer is one trillionth (1/1,000,000,000,000) of a metre.

        The sheets were roughly 5 nanometers in diameter. Graphene is a form of carbon.

        Google tells me that 5 nanometers = 5000 picometers. Is my math off? It seems like there is a factor of 10 between how thick this stuff is and how thick Carbon is.

        • Re:10 Atoms thick? (Score:4, Informative)

          by tylerni7 (944579) on Thursday December 18, 2008 @09:08PM (#26168021) Homepage
          Your math is correct, your chemistry isn't.
          A carbon atom has a covalent radius of about 80pm, but the atoms in sheets of graphite aren't bonded together. I don't know how far apart the atoms would rest, but it's going to be much farther than they would bond.
          • Re:10 Atoms thick? (Score:4, Informative)

            by Chris Burke (6130) on Thursday December 18, 2008 @09:21PM (#26168127) Homepage

            Not that much farther apart, since the article says that the sheets are less than 1nm thick.

            The figure he's quoting is a diameter, which would be the 2d dimensions of the sheet on the surface of the silicon they grew it on. It's the 5nm diameter that makes this exciting as a memory technology since that is very dense.

            • Re: (Score:3, Interesting)

              by hairyfeet (841228)

              I too read TFA and it seems to me they are leaving out the most important part(unless I missed it, which is: How many read/writes can they get out of this before it is toast? Because it can be the smallest, toughest little chip in the world but if you only get a couple of dozen read/writes out of it before it is toast than it'll be pretty damned useless. current read/write for NAND flash is up to,what, 1 million? So at the very least they'll need to shoot for that, and if you want to use it in space explora

              • Re: (Score:3, Informative)

                by elashish14 (1302231)
                In case you didn't RTFA, NAND technology is predicted to reach its size limit in 2012 at 20nm. Graphene can reach much smaller than that. Additionally, they mentioned that it can already run at 100ns (read speed I assume) whereas MLC (current SSD bleeding edge) reads at 50ns right now.

                The current things that are holding it back right now are probably mass distribution and reliability. Honestly though, it will take a lot more to convince me that we'll be using graphene-based memory chips someday.
                • by hairyfeet (841228)
                  Why exactly do we NEED flash to get any smaller? So we can snort flash like lines of coke? Seriously we have multi-Gb MP3 players now that are smaller than my pinkie finger, and there are plenty of 4, 8, 16Gb+ flash "sticks" that are so damned tiny they could easily get lost in the couch or sucked up into the vacuum, so why exactly do we need them to get any smaller? And I did readt TFA but my question didn't have anything to do with size. My question concerned read/writes which IMHO is a whole hell of a lo
                  • by cnettel (836611)
                    If you shrink the size by a factor of 10 and are worried by the number of write, just do wear levelling over a memory ten times the size. As long as it hasn't failed, and you also don't care about speed, write it redundantly with loads of ECC. This is possible if the memory is damn small and damn cheap.
                  • by nasch (598556)

                    There are always applications found when something gets smaller (physically), bigger (storage), faster, or cheaper. It will be the same with this. When we reach the limit of storage density for flash memory, people will still want and expect bigger drives (capacity), but they won't want them to be physically bigger. Thus, we'll need some other technology to satisfy that market. There will be money in it, so somebody will find a way, whether it's graphene or something else.

            • 80 pm vs. 1nm isn't a big difference? 1nm=1000pm, and 1000/80= 12.5X. That's an order of magnitude, and on the quantum scale, that's huge.
              • Re: (Score:3, Informative)

                by Chris Burke (6130)

                There are 10 atoms, so that's 800pm, which is close to 1nm yeah. :)

                Which, uh, you figured out to much greater accuracy than I know how to in another post. Hehe.

                • No! there aren't 10 atoms! I said in the other post that the bonding between graphene and some other surface parallel to the plane of the atoms is >>200 pm. It makes sense for a bond to be this thick, especially since the pi* orbital is very poor at overlapping (which is favorable for bonding).
                  • by Chris Burke (6130)

                    Whatever you say, I have no idea, I suck at chemistry. All I know is the article says it's 10 atoms thick. :)

        • Re:10 Atoms thick? (Score:5, Informative)

          by Chris Burke (6130) on Thursday December 18, 2008 @09:18PM (#26168105) Homepage

          The sheets were roughly 5 nanometers in diameter. Graphene is a form of carbon.

          Google tells me that 5 nanometers = 5000 picometers. Is my math off? It seems like there is a factor of 10 between how thick this stuff is and how thick Carbon is.

          One is talking about thickness, the other a diameter. The next paragraph of the article it says the sheets are a little under 1nm thick, and 10 C atoms would be around 800pm so that's a little under 1nm. The 5nm diameter would then be the other dimensions, these grown sheets are presumably circular. That dimension is important because that indicates how densely you could pack them on a surface.

          • Re: (Score:3, Insightful)

            by Anonymous Coward

            Thank you, that makes much more sense. I think I've got it now. Let me try explaining it with a holiday metaphore:

            What they have created is, say, like a cookie. Each of these little cookies are 5 nanometeres in diameter. It's important to know that, because it lets us know how many cookies we can fit on our cookie pan. Each of these cookies are about 1nm tall. This is important because it affects how many of these cookie trays we could stack on top of each other in the oven.

            I was having a problem conceptual

          • Re: (Score:3, Interesting)

            by elashish14 (1302231)
            Graphene is an array of sp2 hybridized carbon, meaning the HOMO is the pi bonding orbital, and the LUMO is the pi* orbital. The average electronic radius in the p orbital is a bit under 4 times the Bohr radius = 4*53 pm ~ 200 pm and it's safe to assume that the average distance of the pi bonding orbital is close. Since bonding must take place in the higher energy pi* orbital, it must be >>200pm. 1000pm sounds about right.

            The math isn't hard, but I have to take a shit so I can't do it right now.
        • Re: (Score:2, Informative)

          by dakup (1423621)
          they vaporize it on another material. sorry.
    • Re: (Score:3, Interesting)

      by treeves (963993)
      You'll notice from the data you found that they vary by less than one order of magnitude so it's still a useful approximate measure. Other "measurements" vary as well, for example "floors" to measure the height of a building, "blocks" to measure distance in a city or town, "car lengths" to measure tailgating, "gnat's asses"...oh never mind.
    • by fucket (1256188)
      It's like a double-wide mobile home; It's pretty obvious that the "double" refers to width of one mobile home.
    • by Strep (956749)
      Um. They said Graphite, so we can assume it's a carbon atom.
    • by kyc (984418)

      Just to give an idea for us to see the colossal difference between the 'everyday experience' and the atomic world.

      There's really no difference between Hydrogen (Z=1) and Ununoctium (Z=118) when you peek at them from a dimension that is 10 billion times (say the inter-atomic distance is about 1 Angstroms and we live in the meters range) larger than those.

  • my Lisp program dies?

    '(one two three four five six seven eight nine ten ...?)

    *duck*

    • Re: (Score:2, Funny)

      by Anonymous Coward

      You're doing it wrong, you should have counted from zero.

  • by bugnuts (94678) on Thursday December 18, 2008 @08:01PM (#26167443) Journal

    You slide these right off the end of your pencil onto paper.

    You know, pencils make pretty good r/w memory, too, although the number of r/w cycles is limited.

    • by jmerlin (1010641) on Thursday December 18, 2008 @08:03PM (#26167475)
      Who knew? The most advanced memory created yet was invented far before the computer...
      • Re: (Score:3, Funny)

        by revoldub (1425465)
        Who would have thought, thousands of years later, thousands of advancements in technology, and we're back to writing on rocks.
      • Yes, and they're called stone tablets. Luckily, computer researchers seem to be picking up on this, now that they're using graphite. Unfortunately it sounds like "etched in stone" will soon mean "subject to formatting".

    • Re: (Score:2, Insightful)

      by Anonymous Coward

      You know, pencils make pretty good r/w memory, too, although the number of r/w cycles is limited.

      Your comment is clearly funny, but I wonder how these last compared to other forms of graphite.

      The article doesn't seem to mention anything about this memory's reliability or wear -- even theoretical stuff would be fine considering that the technology is relatively new.

    • by quenda (644621)

      You know, pencils make pretty good r/w memory

      You can prototype this new technology at home. All you need is a 4000H pencil, a laboratory-grade pencil sharpener, a microscope, and a steady hand.

    • Re: (Score:2, Insightful)

      by Anonymous Coward

      You know, pencils make pretty good r/w memory, too, although the number of r/w cycles is limited.

      Please explain to me how my pencil can do the read part of r/w memory.

      • by Anonymous Coward on Thursday December 18, 2008 @11:39PM (#26168963)

        You know, pencils make pretty good r/w memory, too, although the number of r/w cycles is limited.

        Please explain to me how my pencil can do the read part of r/w memory.

        Well look at you, you're all the fun at parties, aren't you?

      • Please explain to me how my pencil can do the read part of r/w memory.

        Well, if it's that hard and sharp, you could electrify the end and read the charge differences as you move in a raster pattern, moving across atoms and atom-free zones on a substrate layer. Try it by writing "IBM" on silicon in individual atoms, then using the same method to scan the area. Would probably be a destructive read, but you could probably do it. You could keep the excess atoms in a bit bucket.

        NERD = Nerd Emulating Recursive Datum

        • by rcw-home (122017)

          Try it by writing "IBM" on silicon in individual atoms, then using the same method to scan the area

          "Details of the implementation have been left as an exercise to the reader" isn't a particularly good way to get R&D funding.

          • Try it by writing "IBM" on silicon in individual atoms, then using the same method to scan the area

            "Details of the implementation have been left as an exercise to the reader" isn't a particularly good way to get R&D funding.

            (sigh) sorry, this was done long ago by IBM labs using a scanning tunnelling microscope. http://www-03.ibm.com/press/us/en/pressrelease/20360.wss/ [ibm.com] I thought that was fairly well known by now.

            For the rest of you, apologies for the explanation.

      • The written part will be slightly smoother. Slide the pencil over the part written on and analyse drag for a destructive read. Alternatively, pick up and drop the pencil a very short distance and measure the shock.
  • So.... (Score:5, Funny)

    by Antony-Kyre (807195) on Thursday December 18, 2008 @08:02PM (#26167449)

    no more microwaving your hard drive to aid in data destruction.

    • Re:So.... (Score:5, Funny)

      by Firehed (942385) on Thursday December 18, 2008 @08:33PM (#26167713) Homepage

      I'm pretty sure that microwaving your hard drive only aids in microwave destruction.

      • It all depends on how you do it. If you put the HDD in with the electronics facing up, a 10 second burst will ensure that not a single semi-conductor or chip on the board will ever operate again. No amount of extra microwaving will destroy the data on the platters unless you open up the drive. Then, you might as well use a hammer. Your graphene memory chip would also be a smoking mess after 10 seconds of microwaving.
        • after 10 seconds of microwaving.

          Overkill; five or six seconds are almost always enough to make a smoking mess out of most anything worth putting in there, but only when the item in question should never be microwaved in the first place(tm).

          Which is to say; "food (and so on) should be taken to, at least eleven(tm) , (..and possibly beyond.)

          - You're welcome.

        • No amount of hammertime will completely erase a HDD platter. I could imagine that if surface plamsons occur on the platter while exposed to 700 watts of destruction, the EM field gradients would be so strong that the sticky bits would just evaporate. Microwave radiation isn't high-frequency enough to directly cause this, but with all that power you're damn sure to do more damage than a hammer will

          And opening a HDD is trivial, as all geeks know. :)

  • by gluefish (899099) on Thursday December 18, 2008 @08:03PM (#26167463)
    The problem with using Graphene for write-only memory is that you need Pink Latexene to delete it. Fortunately they've discovered how to make extremely tiny cylinders of Pink Latexene, mounted on the end of yellow wooden sticks, to do such work. The combination of the graphene on one end of the stick and the pink cylinder on the other promises to allow nearly unlimited read-write capabilities, for mere pennies, distributed easily worldwide.
    • by famebait (450028)

      The combination of the graphene on one end of the stick and the pink cylinder on the other promises to allow nearly unlimited read-write capabilities

      Are you sure you don't mean "unlimited write-delete capabilities"? If you start with write-only memory, and then add the ability to delete it, you still can't read it.

      See http://en.wikipedia.org/wiki/Write_Only_Memory [wikipedia.org]

    • by rubycodez (864176)

      I have even heard of a company called Enron that employed a bulk-erase mechanism to vastly speed up the delete cycle, it was called the Arthur Anderson, but it self-destructed

  • by graft (556969)
    For those who missed it, since it's not linked, a relevant story about researchers creating atom-thick graphene balloons that can hold several atmospheres of pressure. Made with Scotch tape. Yowza! http://science.slashdot.org/article.pl?sid=08/08/08/192227&from=rss [slashdot.org]
    • Re: (Score:3, Interesting)

      by oasisbob (460665)

      I've worked with graphite before in a lab (we used it as a substrate for STM [wikipedia.org].

      Using scotch tape to pull up layers of graphite must be a common technique: we used it too. There are many kinds of graphite. Using crystalline graphite (found in nature), you could use the tape to pull up a nice thin layer.

      Being around improvised solutions using common materials was one of my favorite things about lab work.

  • Vaporware (Score:5, Funny)

    by BlackSabbath (118110) on Thursday December 18, 2008 @08:13PM (#26167545) Homepage

    "...we grow it from the vapor phase..."

    Literally, vaporware.

    • by osu-neko (2604)
      By the time this new storage technology becomes practical, they'll be able to distribute Duke Nukem Forever on it...
  • Phew! (Score:2, Funny)

    Thank god I didn't invest in SSD. Those are so obsolete.
  • by Anonymous Coward on Thursday December 18, 2008 @08:30PM (#26167695)

    the RIAA et al will be wanting royalties off every pencil sold and Canada will have a pencil tax?

  • by psnyder (1326089) on Thursday December 18, 2008 @08:32PM (#26167703)

    Reading the articles, it appears the size is nice, but it isn't the biggest deal here. They're projecting a bit smaller than 10nm, which is twice as small as next-generation flash drives that "projections show ... will reach its limit of 20nm by around 2012."

    The biggest deal here seems to be power management.

    What distinguishes graphene from other next-generation memories is the on-off power ratio - the amount of juice a circuit holds when it's on, as opposed to off. "It's huge - a million-to-one," said Tour. "Phase change memory, the other thing the industry is considering, runs at 10-to-1. That means the 'off' state holds, say, one-tenth the amount of electrical current than the 'on' state."

    Current tends to leak from an "off" that's holding a charge. "That means in a 10-by-10 grid, 10 'offs' would leak enough to look like they were 'on.' With our method, it would take a million 'offs' in a line to look like 'on,'" he said. "So this is big. It allows us to make a much larger array."

    • by Chris Burke (6130)

      Current tends to leak from an "off" that's holding a charge. "That means in a 10-by-10 grid, 10 'offs' would leak enough to look like they were 'on.' With our method, it would take a million 'offs' in a line to look like 'on,'" he said. "So this is big. It allows us to make a much larger array."

      Er, I'm don't get that, since if this is going to be memory then you have to account for the fact that it's possible every single bit could be a 1. And current certainly leaks from the higher-voltage 'on' state.

  • Interesting. That is how artificial diamonds are formed too... vapor forming around a diamond seed in a vacuum chamber.
    • Interesting. That is how artificial diamonds are formed too... vapor forming around a diamond seed in a vacuum chamber.

      Chemical vapor deposition. I don't think it's in a vacuum though. A vacuum (by definition) is an absense of matter. Chemical vapour deposition works (IIRC) by having a gas (such as methane... i.e. matter) which is heated and then doing magic to seperate carbon from diamond deposits.

  • Graphene/Graphite (Score:5, Informative)

    by kyc (984418) on Thursday December 18, 2008 @11:42PM (#26169001)

    Graphene has been studied extensively in the last few years. Carbon Nanotubes were on the rise (which are just rolled up sheets of single layer graphite) but the current difficulties to manipulate those to create devices staggered their advance. Graphene ( or Graphite for that matter) is a little easier to manage because it's like a 2 -D sheet and it can be laid/printed off a substrate more easily.

    The current major problem of graphene is the lack of a sizable band-gap which is typically required for semiconductor modulation. We may see a breakthrough in the following years if people figure out a way to overcome this barrier.

  • Finally, memory you can erase.
  • I wonder if that means I can get a heap of this and create a stack [wikipedia.org] now...

    GrpA

  • withstand temperatures of 200 degrees Celsius and radiation that would make solid-state disk memory disintegrate

    Securely destroying such a drive before disposing of it may be a challenge...

    • Re: (Score:3, Funny)

      by 3.5 stripes (578410)

      Not really, you'll just need to try and take a very important test with it.. it'll break almost immediately..

    • Not really, you just heat it over 200 degrees celcius, you can get that on most home ovens, you just by burning it.

    • Securely destroying such a drive before disposing of it may be a challenge...

      Really? 200 degrees C is only 392 degrees F.

  • ...after reading the article I'm still wondering what the storage mechanism is.

    OK, so it's made of graphene, but how does it remember anything, and how do you read and write it? It's like launching a new kind of engine, and only specifying what it's made of.

    Anyone got a more detailed link?

    We are actually given a lot more hints about how the main competitor works, albeit only by virtue of its name, not journalistic thoroughness.

    • Well, my bet would go to the memory being NAND FLASH, with graphene transistors. But I also can't find it anywhere.
  • Get over it. There will be no faster computers now that the US Govt has bailed out the DRAM industry. Innovation like this is illegal!

Money will say more in one moment than the most eloquent lover can in years.

Working...