DIY Solar Panels


Reader [unangst] pointed out to us an article in the U.K.’s Daily Mail, where a teenager from Nepal had managed to create a 9v, 18W solar panel using human hair rather than the usual semiconductors (usually crystalline-silicon). The complex silicon in solar panels are what keep the prices out of reach of developing nations, and while there are a number of new technologies that are helping  bring down the cost, [Karki] managed to make his solar panel for only £23 (roughly $38). He also claims that when mass produced the price could drop substantially down to under $10 a panel, which would shatter the $1/watt sweet spot.

The melanin in hair acts as an organic-semiconductor, and while the hair does not have the longevity that silicon panels have (months rather than years), these panels can be made cheaply and serviced with little to no complex knowledge. Using melanin as an organic semiconductor seems to be a newer idea, because information seems hard to come by, but we managed to find a research paper from 2007 that explored the energy absorption attributes of melanin, as well as some good background info for the science types.

Research Paper (Warning: PDF)

So, Hack a Day readers, which one of you is going to make your home-brew solar panels first? Let us know when you do.

Thanks [unangst].


74 thoughts on “DIY Solar Panels

  1. seriously? we’re falling for this? If we leave aside the issue of there being no actual explanation of how this works, the surface area of the hair in this system is so small that the amount of solar energy falling on that area would be tiny, and it would be impossible to produce the amount of power they’re claiming

  2. i dont think they mean the person uses hair as the actual semi conductor but rather the melanin in the hair so his panel isn’t made of hair but of the melanin and the news people were just trying to compare it to somthing

  3. I admit, it’s got me a bit suspicious, the Daily Mail is not one of the best papers in the UK; not as low as an outright tabloid, but nothing like The Guardian.

    The paper that hackaday found supports the theory. Unfortunately, is a PRE-publication site, there are no comments, and only four votes for the paper, so it hasn’t been seriously reviewed.

  4. Not possible given the size of the panel and “density” of hair as shown in the photographs. The actual surface area of hair receiving illumination would not provide 18W even at 100% conversion efficiency. Joke, or Scam… (Sigh.)

  5. While melanin may have important applications in photochemistry, what these guys are doing is pretty fake. As someone has pointed out, the surface area of the hair is insufficient to produce the power output claimed – particularly as the background is black and will not reflect light back onto the hair.

    Secondly, hair is a pretty darned good insulator. Try it with your multimeter if you don’t believe me. gotta call “bullshit” on this one.

  6. Note in the artical that two dissimilar metals – copper and aluminum:

    “We proceeded to test the photo electrochemical properties of melanin by
    manufacturing a prototype cell (figure 2). We started with a very simple cell were
    the electrolyte was a 1.3% solution of melanin in distilled water, cooper and
    aluminum electrodes 2.5 cm apart, cooper wires (covered with silicon) where
    attach to the electrodes by glued them down, we noticed that any kind of
    welding affected the melanin’s behavior. The cell started to give up electricity
    just a few minutes after being ensemble.
    and that the output declined after a few hours”
    – sounds more like a battery to me…

  7. Possible, energy is max 1000w per sq meter, by looking at the pictures in the article, the hair is used to colour the back panel, the lumps and what seem to be hairs are the front connection wires and not collection devices…

  8. Kevin: The ars tehnica article is pretty bad contradicting itself several times and messing up GB and MB. I wouldn’t trust it anymore than the original research.

    This “sounds” fake but I think they’re using hair as a cable only and everybody is just misunderstanding

  9. I’m not denying that organic solar cells work, or even that they might use melanin as the active dye, but there is also the fact that the panel shown looks to be about the same size as an 18W silicon panel (which is a pretty substantial size), even though “state of the art” organic cells are still far below silicon in efficiency. (there main point of interest is vastly cheaper manufacturing.) Organic melanin-based solar cells? Maybe. 18W from an amateur-constructed panel of the size shown? No way.

  10. Seriously hackaday, get a grip. Even is this weren’t bunk, your story is riddled with problems — $1/W has already been broken, longevity is one of THE major factors in the cost structure of cells, and developing nations are one of the major benefactors of the solar industry since they can largely bypass the need to scale an electrical infrastructure based on centralized production. Boo.

  11. I just wanted to comment on the “comment” about the surface area. I don’t know if you guys know but I did read an article where the researches ware using a “fuzzy” matterial that resulted in greater efficiency^2 … however I’d like to have a panel that didn’t need an haircut every few weeks, I forget to cut my already and I don’t need anther excuse for the home owner association to bug me with. “Sir would you please give your roof pannels a trim” …

    … passer by “get your house a haircut hipie!”

  12. The nature link is complete bs. PDFs on nature precedings are not peer-reviewed at all–they likely did this to make themselves look legitmate. I’m pretty sure they measured something wrong.

  13. O.K. Here is a stupid question. Can anyone tell me why the ‘scope’ display in the pdf figure is showing an AC waveform? Shouldn’t the output from a ‘solar cell’ be DC.

    Also, the LED has been lit for 10,000 hours? Thats over a year of contiuous operation – so they had this working a year ago and are now just publishing this paper.

    If this is legit then it is very interesting. However, I have serious doubts after reading the paper.

    I actually thought for a moment to see if it was April fool day – just a few months off.

  14. no, no, no… nobody’s going to be homebrewing this. this will only matter if someone figures out they have to first invest in some cheap easy-to-set-up permanent infrastructure, so they can change out the very-mass-produced panels every month or whatever… and they have to arrive via a bunch of shipping just-in-time because nobody’s going to want to sit on them while they race toward expiry… or they have to be preservable while stored somehow at the very least… i need to know more about it before i can consider it even partially hopeful anywhere.

  15. @ambigrid review: This student is making a solar *cell* not a solar panel. The student’s technology is based on cuprous oxide which provides about 0.25V at 50uA per 0.01 square meter. To generate 9V at 18W this student’s solar panel would need to be 120m x 120m square. I assume that you mean you assembled a solar panel from silicon cells, correct? If not silicon, what technology is used in the cells?

  16. @Dan Fruzzetti: I agree. Maintenance is going to be an issue for a cell made from human hair. As described in an online article [ ] this is a *wet* cell, meaning the hair has to be saturated with a salt water solution to work. I am not sure what it will happen if you keep a mass of hair wet for a month or two, but I am guessing it won’t be pretty. So it’s not just a matter of swapping out the hair, the hair has to be kept wet for the cell to work. I’d be interested in having as many expert eyeballs as possible go over my debunking site, so if you have time, please take a look:

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.