Technology Review - Published By MIT
Advertisement

More Efficient, and Cheaper, Solar Cells

New manufacturing techniques could cut solar power costs by 20 percent.

By Kevin Bullis

Monday, September 14, 2009

smaller text tool iconmedium text tool iconlarger text tool icon

Improvements to conventional solar cell manufacturing that could significantly increase the efficiency of multicrystalline silicon cells and bring down the cost of solar power by about 20 percent have been announced by startup 1366 Technologies of Lexington, MA.

Light trap: Incoming light reflects off grooves in a silver band and is redirected along a glass cover. This light, which is usually lost, can then be absorbed by the solar cell. The grooved band is one of three improvements that could significantly lower the cost of making solar power.
Credit: Technology Review
Multimedia
video  Take a tour of the labs and pilot manufacturing line at 1366 Technologies.
video  Watch a detailed demonstration of how to make solar cells more efficient.
video  See how solar has changed in the last 30 years, and where it's headed.

Such cost reduction would make solar power more competitive with conventional sources of electricity. In sunny environments, this could bring the cost of solar down to about 15 or 16 cents per kilowatt hour, says Craig Lund, 1366 Technologies's director of business development. That's cheaper than some conventional sources of electricity, especially those used during times of peak electricity demand.

1366 Technologies has developed three processes that can be incorporated into existing solar cell manufacturing lines to improve cell efficiency. It has shown that these technologies can be used to produce multicrystalline solar cells that are 18 percent efficient at converting sunlight into electricity. The current industry standard for such solar cells is 15 percent to 16 percent, according to Joonki Song, a partner with Photon Consulting, based in Boston, MA, although higher efficiencies have been reported. The company has demonstrated the new technologies before, but only with very small, experimental solar cells in a laboratory. Now it's made full-size solar cells using the type of equipment used in large-scale manufacturing.

The key to the startup's technologies, however, isn't the efficiency that it's achieved, but how little that efficiency costs. Lund says that the new processes add only a few cents per watt to the cost of fabricating solar cells, but this investment leads to much greater cost savings in the final product. Improving the amount of power each solar cell generates lowers materials costs, solar module manufacturing costs (in which cells are assembled into solar panels), and installation costs. In the end, Lund says, the cost of an installed solar panel will be reduced by 50 cents to 80 cents per watt.

Story continues below


The new processes, which were invented by Emanuel Sachs, the company's chief technology officer and a professor of mechanical engineering at MIT, all increase the amount of light that solar cells can absorb.

In a normal silicon solar cell, electrons generated in the silicon must make their way out of the material to produce an electrical current, traveling first to the top layer of the silicon and then along this layer to narrow silver lines called "fingers." The fingers then conduct the electrons to the busbars, two or three prominent silver bands seen on the surface of most silicon solar cells. These bands shade the silicon under them, reducing the amount of light the cells can absorb.

Comments

  • Cheaper solar panels
    This is quite an achievement if the panels perform as presented. I cannot wait to see photovoltaic solar panels drop in price to compete with current forms of energy without rebates. With research and investment such as this, the day is coming.
    Rate this comment: 12345

    clean family...
    09/14/2009
    Posts:1
    Avg Rating:
    4/5
  • $ / kW; ¢ / kWh; 24 x 365
    GO, GO, GO!

    Getting there.
    Solar doesn't work at night.
    It requires about 4x the kW *and* storage.  (Solar only gets, at best, ~2000 hours of operating time out of the 8760 hours per year).

    But doesn't beat coal, yet.  :(
    (~@$2.20/W; ~5¢/kWh, 24x365)

    Solar needs to get to ~50¢/W.
    ***Including*** the inverter.  :(
    Which would mean the panels have to get to under ~25¢/W?
    At least, to replace coal.  :)
    Rate this comment: 12345

    nekote
    09/14/2009
    Posts:139
    Avg Rating:
    3/5
    • Re: $ / kW; ¢ / kWh; 24 x 365
      It's not just about the bean counting.  Given the choice between solar or coal, people would opt for solar if the price was similar.  I don't think that solar has to significantly undercut coal for it to take off.  But when the price does finally undercut it, that will be a great day for humanity.
      Rate this comment: 12345

      jpm1u
      09/17/2009
      Posts:13
      Avg Rating:
      4/5
  • Costs...
    In the USA the average KHW cost is 9.3c so there are very few people paying 5c...

    Secondly the 5 hour figure you are quoting is the average number of Suns for the USA. Which is the amoung of 1000w falling on one square metre of surface averaged over a year.

    Now I have a 2.1kw system at home and it runs at 3kwh a day in the winter months and 14kwh in the summer months so I get about 7.5kwh a day on average.

    The drop mentioned in the article was to push the price down to 80c a watt. My understanding is that Solar PV becomes viable at around a dollar a watt for competing against coal.

    I purchased my panels at around the $10 a watt. The price has dropped and is around $6 a watt. That's local prices.

    I havent had a power bill in some time and on the whole I think its working nicely. As for coal, if the true price comes into play then start looking at rapid increases in the cost of power.

    The years of wasting energy are drawing to an end.

    Solar thermal and using reverse water storage pumping to store peak solar power for 24 by 7 use is all possible.

    Coal wont go over night but more of the power requirements will be met by other means and a reduction in energy use as the costs go up for carbon based power.
    Rate this comment: 12345

    Thaxone
    09/15/2009
    Posts:4
    Avg Rating:
    5/5
  • Competing with coal
    Of course, $/W is only an approximation.  Different types of solar cells vary widely in their installation costs and generate different amounts of kWh for their ratings (thin film does better in the heat, for example).

    We've a long way to go before solar has to start displacing coal at night. When we get to that point, the big challenge will be energy storage--we may need some fundamental breakthroughs to make that cheap enough for us to rely on solar. It could be we need to keep burning coal, while capturing the carbon dioxide. Or use more nuclear and geothermal.

    Check the third video above for Sachs take on this issue.
    Rate this comment: 12345

    Kevin Bullis
    09/15/2009
    Posts:101
    Avg Rating:
    4/5
  • [no subject]
    good article and good interviews with the 1366 tech guys, thanks.
    Rate this comment: 12345

    jpm1u
    09/15/2009
    Posts:13
    Avg Rating:
    4/5
  • No foclear
    We do not need more foclear (FOssil fuel + nuCLEAR) energy, only renewables. Regards.
    Rate this comment: 12345

    NoFoclear
    09/19/2009
    Posts:1
    Avg Rating:
    3/5
  • [no subject]
    Best way to go. There are lots of ways to improve 3-D nanopillar photovoltaics for higher performance, and ways to simplify the fabrication process as well. This is promissing.
    Rate this comment: 12345

    jmongu
    10/01/2009
    Posts:5
    Avg Rating:
    2/5
  • Halftone photo resist to print hexagonal anti-reflection surface?
    Dr. Sachs was coy during the interview regarding the mask process used to etch out the honeycomb anti-reflection pattern; however, halftone masks would probably work fine. See for instance "Fabrication of complex micro-optic components using photo-sculpting through halftone transmission masks", D R Purdy, Pure Appl. Opt. 3 (1994) 167-175.
    Rate this comment: 12345

    mwarren
    10/15/2009
    Posts:1
    Avg Rating:
    1/5

Log In

Forgot your password?     Register »
Advertisement

Videos

Malleable Maps, Artistic Robots and Bubble Interfaces
Technology Review January/February 2010

Current Issue

Security in the Ether
Information technology's next grand challenge will be to secure the cloud--and prove we can trust it.
Advertisement
Advertisement
Advertisement
Subscribe to Technology Review's daily e-mail update. Enter your e-mail address

TECHNOLOGY RESOURCES

More Technology News from Forbes

Advertisement
MIT Massachusetts Institute of Technology © 2010 Technology Review. All Rights Reserved.