Skip to main content

View Diary: Nearly triple efficiency and lower cost: thin-film solar cell breakthrough (281 comments)

Comment Preferences

  •  For the grid (18+ / 0-)
    By eliminating the expensive ITO layer, a thin-film cell made this way is cheaper than current thin-film technology. (Chu used gold for his mesh, because it's easy to manipulate in the lab, but it could just as easily be made from aluminum.) Fabrication of the grid is extremely easy. You start by creating a cell-sized mold using light interference techniques, then use the mold to run off any number of grids you want.
    I don't think you can just swap the gold with aluminum. Al will form a thin oxide layer that will kill its conductivity and thus efficiency. Because it's exposed to the elements you really need a noble metal like gold or silver to do this I would fear. The silver used in the top contact of silicon panels continues to be a pain point for makers but they just can't find anything better.
    •  Couldn't an anti-reflective coating be applied (10+ / 0-)

      above the AL?  The purpose of the grid isn't to leverage the raw conductivity of the metal, but to realize a plasmonic effect.  So depending on the design, it might still work well even in the presence of oxidation.

      •  Plasmonic effects are material-dependent (16+ / 0-)

        Miggles is right that the grid's conductivity isn't anywhere near as important as the grid's plasmonic properties--the short version of which is the wavelengths of light that the grid absorbs or scatters.  Those absorption and scattering properties change somewhat with the size and shape of nanomaterials, but they change a lot when going from one element to another--even if the nanomaterials have perfectly identical dimensions and/or patterning.

        Take the case of plasmonic nanoparticles, for example, which my lab uses on a daily basis in our research.  A lot of different elements show plasmonic absorption and scattering--gold, silver, platinum, copper, aluminum, gallium, etc., etc., etc.  Each of these elements has what's called a "native plasmon", which is basically the wavelength at which the material most effectively absorbs and scatters light.  Gold's native plasmon is in the green part of the spectrum, around 520 nanometers.  Copper and platinum have native plasmons further toward red and infra-red wavelengths, silver's native plasmon is in the blue part of the spectrum (around 420 nanometers), and aluminum's native plasmon is down in the ultraviolet (below 380 nanometers).  

        The sun's light is strongest in the green and yellow parts of the spectrum, which matches gold's native plasmon really well.  As a result, Chou's gold mesh is very effective at trapping sunlight.  A copper or platinum grid can't trap green light, but will trap light in the red portion of the solar spectrum (copper's very reactive, though, so it's probably a poor choice in Chou's application).  A silver mesh could trap light ranging from the blue part of the spectrum out toward green wavelengths, but it would be nearly impossible to use a single silver mesh to trap all light from the blue through the red (silver's also very reactive, like copper).  With a native plasmon in the ultraviolet, an aluminum grid just isn't well-suited to trapping anything more than ultraviolet light and very high-energy blue light.


        Aluminum and gold interact with sunlight quite differently, so an aluminum mesh (regardless of whether or not there's an oxide layer) won't trap most sunlight the way a gold mesh does.

        You think it's hot? Imagine what it would be like if global warming really existed!

        by JSc on Fri Dec 07, 2012 at 10:50:59 AM PST

        [ Parent ]

        •  Thanks very much for this (4+ / 0-)
          Recommended by:
          FarWestGirl, Miggles, elwior, indie17

          You learn something new every day, and you just taught me something.

          We are all in the same boat on a stormy sea, and we owe each other a terrible loyalty. -- G.K. Chesterton

          by Keith Pickering on Fri Dec 07, 2012 at 03:29:40 PM PST

          [ Parent ]

        •  All irrelevant. :) (7+ / 0-)

          Even if they have to use gold, at 30nm, it'd take 1725 square meters  to use one kilogram of gold), or about 9 cents per nominal solar cell watt at the current $40k/kg price of gold and ~25% panel efficiency.  Which is nothing.

          A couple general comments about the article.

          1) Geez, is anyone else doing groundbreaking cleantech research out there except Yi Cui?  ;)  Seems like 80% of the awesome discoveries out there in batteries and solar come from his team.

          2) "Yeah, I know: every week, somebody somewhere announces some sort of solar cell "breakthrough" that turns out to be ... well, not much. I see them all the time too." - actually, I find that sentiment just as bad as the sentiment it's criticizing.  The reality is that, yes, most new techs don't play out at the commercial scale.  But a lot of them do.  You just never hear about when they make it into commercial products, so as far as you know, nothing happens.  For example, go back to DK from several years ago and you'll see articles aabout the use of silicon electrodes in lithium-ion batteries.  It was one of those, "wow, amazing!" type stories that the author just dismissed.  Well, silicon electrodes now are in some of the newest types of lithium-ion batteries.  Not the most advanced form, mind you, but a serious first-generation with huge potential for continuing evolution.

        •  at $1800/oz (0+ / 0-)

          you aren't going to convince anyone that this is commercializable

          •  Do the math (7+ / 0-)

            One square meter of this stuff (30 nm thick gold, with holes) would weigh about 0.007 ounces. At $1800 per ounce, that's $13.36 for enough surface to generate 80 Watts (at 8% efficiency, and that's before most mesh parameters have been optimized). Or 17 cents per Watt.

            Further, there may well be other materials with similar properties that mimic gold. A copper/zinc alloy might work just as well or perhaps better. There are a lot of things to try.

            We are all in the same boat on a stormy sea, and we owe each other a terrible loyalty. -- G.K. Chesterton

            by Keith Pickering on Fri Dec 07, 2012 at 04:15:39 PM PST

            [ Parent ]

    •  Silver is noble? (3+ / 0-)
      Recommended by:
      isabel, Lujane, CwV

      I know it's expensive, but isn't it even more reactive than Al?

    •  Aluminum would work fine (16+ / 0-)

      Because in this case it is used as a metal mask not a conductor.

      BTW, a whole heck of a lot (billions and billions) of Integrated Circuits use aluminum film metallization for conductors and even aluminum or copper wires (although gold wire bonding in more common) and it's not really difficult as it seems since oxidation of aluminum is somewhat self-limiting and passivation coatings can be applied.

      Al and Au are very common scientific mirror coatings and Al works best because (a) it is pure white and reflects full spectrum and (b) the back side oxidizes and passivates.

      And please note noble metals also oxidize, see the discussion about silver oxidation (tarnish) elsewhere. This oxidation is just a few molecular layers thick and although it does slightly raise contact resistance as you suggest, it has negligible effect on bulk properties, and the same is true of Aluminum. Silver make great conductors for some purposes, particularly wiper-brush switches such as precision potentiometers (better than gold because the bulk conductivity is lower even if the contact resistance is slightly higher).

      What about my Daughter's future?

      by koNko on Fri Dec 07, 2012 at 08:33:28 AM PST

      [ Parent ]

      •  Clarification (4+ / 0-)
        Recommended by:
        antooo, walkshills, FarWestGirl, indie17

        In fact, the contact resistance of pure silver is lower than gold, what I was suggesting in the above was the case where a silver surface is heavily tarnished due (usually) to the effect of atmospheric sulfur that will catalyze growth of tarnish to several molecular layers, but in practice, silver brush contacts usually remain clean and so the oxidation is a monolayer and not an issue.

        Got to be careful around here, there could be some hot-tempered audiophiles with expensive silver potentiometers reading this thread that would take offense and banish me to hell for even suggesting one would entertain using Gold contact switches (although those folks seem to accept gold plated plug contact surfaces for their bloody precious cables).

        What about my Daughter's future?

        by koNko on Fri Dec 07, 2012 at 09:44:18 AM PST

        [ Parent ]

      •  We used to use SiO, (silicon monoxide), via (2+ / 0-)
        Recommended by:
        indie17, ozsea1

        vacuum deposition to coat astronomical telescope mirrors.

        Worked at Coulter Optical telescopes, many, many moons ago. ::sigh::

        Information is abundant, wisdom is scarce. The Druid

        by FarWestGirl on Fri Dec 07, 2012 at 05:34:05 PM PST

        [ Parent ]

    •  Should be no problem (7+ / 0-)
      Recommended by:
      qofdisks, Lujane, koNko, CwV, walkshills, cai, elwior

      The oxide layer is only one atom thick on aluminum, which is why aluminum is used routinely for wires in electrical applications today.

      We are all in the same boat on a stormy sea, and we owe each other a terrible loyalty. -- G.K. Chesterton

      by Keith Pickering on Fri Dec 07, 2012 at 09:06:59 AM PST

      [ Parent ]

    •  Not a problem in semiconductor (0+ / 0-)

      That aluminum sheet could be electrically grounded through contacts underneath. In fact- this type of operation is fairly standard in semiconductor fabrication whereas operations using gold is quite a bit less common.

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site