Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Networking Technology

Optical Fiber With a Silicon Core 60

Posted by timothy from the now-that-ain't-in-leviticus dept.
Roland Piquepaille writes "According to the Optical Society of America, U.S. researchers have been able to create a practical optical fiber with a silicon core. As they were able to use the same commercial methods that are used to develop all-glass fibers, this might pave the way for future silicon fibers as viable alternatives to glass fibers. The scientists note that this should help increase efficiency and decrease power consumption in computers and other systems that integrate photonic and electronic devices. Here is a good summary by the lead researcher: 'In the past, we've needed one structure to process light and another to carry it. With a silicon fiber, for the first time, we have the ability to greatly enhance the functionality in one fiber.'"
This discussion has been archived. No new comments can be posted.

Optical Fiber With a Silicon Core More

Optical Fiber With a Silicon Core

Comments Filter:

  • Optical + Silicon + The Internet is for...? (Score:3, Funny)

    by Syrente ( 990349 ) on Wednesday October 29, 2008 @02:52PM (#25560017)
    There's a boob joke in here, somewhere.

  • Just How I Like My Women (Score:1, Funny)

    by Anonymous Coward
    ... with a silicon core.

  • Good! Just what the doctor ordered! (Score:4, Funny)

    by Yvan256 ( 722131 ) on Wednesday October 29, 2008 @02:56PM (#25560093) Homepage Journal

    More fiber!

  • How do they do it? (Score:3, Interesting)

    by tsa ( 15680 ) on Wednesday October 29, 2008 @03:08PM (#25560231) Homepage

    OK, I've read TFA. Now all I want to know is: how the hell do they make many many kilometres of crystalline silicon? Amorphous Si, OK, I can understand, but getting it crystalline is amazing. Or do they mean polycrystalline? Wouldn't that ruin the optical properties?

    • Re:How do they do it? (Score:4, Informative)

      by MightyYar ( 622222 ) on Wednesday October 29, 2008 @03:33PM (#25560529)

      The silicon crystals they grow for wafers are huge these days. Presuming that they draw it out the same way that they do traditional fiber optic cable, they'll get plenty of length.

      A BIG optical fiber is 100 microns across (0.1 mm). A big wafer is 300mm. So they stretch down over 3000 times in size, which should give you 9 MILLION times the original length, unless my math fails me. I don't know how long the crystal is that they grow these days for wafers, but one only half a meter long should go a long way at that diameter...

      • Re:How do they do it? (Score:5, Informative)

        by mapsjanhere ( 1130359 ) on Wednesday October 29, 2008 @04:07PM (#25561085)
        the wavers are actually sliced out of big single crystalline cylinders; making the cylinders bigger was what slowed the progress of waver growth. I don't think you can "draw" the crystal in a conventional sense, but since the material is extremely clean, it should recrystalize back into a single crystal out of the melt.

        • Yeah, I meant I wasn't sure how long the single crystal that they slice the wafers from is. They can slice them to about 18 mils - if they want the wafers thinner they have to do an additional step, backgrinding or such. They can get 'em all the way down to 1 mil these days, though the smallest I've seen in production is about 3 times that. Quite flexible at that thickness. At the show, they guy will wrap the wafer around on itself to demonstrate flexibility.

          Am I off-topic yet? :)

      • From the Paper, as referenced in the article, on OpticsExpress [opticsinfobase.org]:

        2.1 Fiber fabrication
        Three tubes of optical quality silica were sleeved concentrically to yield an overall cladding
        with outer diameter about 50 mm and inner diameter of 3.5 mm. A section of bulk silica rod
        was joined to one end of this tube assembly to act as a seal for the silicon core, which would
        be molten during the draw. This approach to layering of concentric tubes was utilized since a
        single glass tube of those dimensions was not commercial

        • So it looks like they start with a silicon crystal in the middle of glass with a silicon crystal end cap. Then they melt the Si in the glass, leaving the endcap intact and do the draw. I suppose at that point they cool the fiber, which would cause a crystal to propagate from the seed (endcap).

          Or did I read that wrong?

    • Re:How do they do it? (Score:4, Insightful)

      by smaddox ( 928261 ) on Wednesday October 29, 2008 @04:53PM (#25561693)

      I was wondering the same thing. Silicon is not a glass, and cannot be stretched like a glass can.

      Glasses have very fuzzy 'melting points'. In other words, they just get gooey rather than becoming a liquid. Silicon on the other hand, has a well defined melting point, and is not gooey.

      Now, silicon crystals ARE grown inside of SiO2 lined graphite crucibles. So, I imagine that they might be able to melt the silicon inside of a Si02 layer, and stretch both. Then, when it cools, the silicon will form microcrystals. So, it wouldn't be crystalline, but it would still be transparent to photons with energy below 1.1 eV (wavelength greater than 1.1 um).

  • ohnoitsroland!! (Score:1, Informative)

    by Smidge207 ( 1278042 )

    I think most of you are aware of the controversy surrounding regular Slashdot article submitter Roland Piquepaille. For those of you who don't know, please allow me to bring forth all the facts.

    Roland Piquepaille has an online journal. . . . It consists almost entirely of content, both text and pictures, taken from reputable news websites and online technical journals. He does give credit to the other websites, but it wasn't always so. Only after many complaints were raised by the Slashdot readership did he

    • Re: (Score:2)

      by gnick ( 1211984 )

      As of today, it is clear that ten articles were accepted in October, six in November, and four in December (so far). See http://slashdot.org/~rpiquepa [slashdot.org] for yourself.

      Umm... What? I count 6 in October and none yet for next month or the month after. Did someone forget to proof-read his copy-pasta?

    • Re: (Score:2, Flamebait)

      by jcnnghm ( 538570 )

      Why do you care so much that he makes a little bit of money for his effort? The stories he posts are usually on topic, interesting, and somewhat obscure. It's a shame the commentary is always clouded by the whiners. The people that continue to bitch about this really should just get over it already.

      • Re: (Score:3, Insightful)

        by Ethanol-fueled ( 1125189 )

        He copies-and-pastes the meat of his journal entries from professional and academic journals and news magazines

        Some mods call this "karma whoring", but this is slashdot, and who wants to navigate through 10 pages just to RTFA when half of the participants don't even read the fuckin' articles? :)

    • So he's an aggregator. So is Google News, so is Slashdot. He makes money at it. Get over it.

    • Re:ohnoitsroland!! (Score:4, Insightful)

      by clem ( 5683 ) on Wednesday October 29, 2008 @04:01PM (#25561009) Homepage

      I don't see how unfair it is that Roland Piquepaille should receive compensation for what effectively is online research. It's at least as fair as Slashdot, a for-profit company, getting loads of free online research from article submitters.

      How do you lose out by his $80 an article?

      • Re: (Score:1, Insightful)

        by Anonymous Coward

        it's perfectly fair. What isn't fair is the number of articles submitted vs accepted ratio, for most slashdotters this is simply 0, roland p. seems to have an inside line that pretty much guarantees acceptance of his articles, no matter how flimsy. To see this in action have a look at the firehose and how much content there is available of better quality and more relevance (and dare I say more original) than rolands drivel, and yet, his gets selected every time.

    • I looked at that blogads.com site and found it extremely hard to navigate, very raw etc... turns out that adblock+ disables the CSS on the site because it containst "ads" in it :-p

      just my 2c

  • will this new silicon fiber be more durable and less fragile than glass-fiber?

    if not, what's the point?

  • Linear computing perhaps? Much more interesting than wearable computing. Reading TFA made me imagine multi-strand silicon core cable, one core being a sort of stacked CPU, another core being RAM, inductive bus (semi-conducting insulation perhaps?) another core power, one a colossal shift register, another a sort of EPROM for archival storage, etc. Why send bits along the wire when the wire could supply the bits directly? Wire archive?

  • Cheaper net for the aussies (Score:3, Insightful)

    by afidel ( 530433 ) on Wednesday October 29, 2008 @09:18PM (#25564451)
    The first thing I think of when I hear increased efficiency is not "oh great now my GBIC's will save a few mW of power" it's "that's great for international communications". The reason is a big part of the cost of an undersea cable is the boosting equipment and the weight and bulk required to provide power to them.

  • Does this mean the Earth is a giant computer powered by lightning? :)

  • ...lies in the accentuated non-linear effects of Silicon over glass. As the article states, frequency-shifting and other optical processing feats can be performed in the fiber instead of having to do the photon-electron-photon dance. This makes WDM [wikipedia.org] signal generation/detection much simpler. Imagine doing all the functions of one of these [wikipedia.org] by choosing the right frequency-shifting fiber. The industry could standardize on a single laser frequency (193.10 THz) and insert DWDM signals by using fibers that shift th

  • from the OpticsExpress paper:
    "measured propagation losses were 4.3 dB/m at 2.936 μm"
    100x more than regular fiber, at wavelengths we don't use for communications... I'm not sure if I see how this is useful.
    I suppose it's cool to have a Process to make Si-core fibers, but it's not like computer chips need 2 km of fibers inside, and it's clearly not useful for C-band optical communication,

    Seems like it's just a buzzword due to "Silicon" being placed nearby "photonics".

    *maybe* you can

    • Re: (Score:1)

      by kooma ( 92065 )

      from the OpticsExpress paper:
      "measured propagation losses were 4.3 dB/m at 2.936 μm"
      100x more than regular fiber, at wavelengths we don't use for communications... I'm not sure if I see how this is useful.

      Actually, it's worse. You failed to note that these guys measure in dB/m, whereas typical fiber losses are in dB/km...

Slashdot Top Deals

Genius is ten percent inspiration and fifty percent capital gains.

Close