Technology Review - Published By MIT
Advertisement

Thin Film's Time in the Sun

First Solar's thin-film technology is now challenging silicon panels at large-scale solar-power facilities.

By Peter Fairley

Friday, July 27, 2007

smaller text tool iconmedium text tool iconlarger text tool icon

The low manufacturing cost of photovoltaics that employ thin films of cadmium-telluride semiconductor have long been seen as having the potential for lifting solar power from its niche status as a very expensive power source, delivering less than a twentieth of 1 percent of U.S. electricity.

Solar blitz: Cadmium-telluride thin-film solar panels made by First Solar are installed on a rooftop in Bamberg, Germany, by Blitzstrom, a German solar-system integrator. Thin-film technology is starting to compete with traditional silicon technology, especially in countries like Germany, where financial incentives are strong.
Credit: Blitzstrom GmbH
Multimedia
•  View images of thin-film solar panels.

Now, after two decades in which cadmium-telluride technology was dogged by low power output and reliability problems, it's suddenly elbowing its way into renewable-energy markets and competing with today's dominant solar technology: silicon solar panels. The company behind this technology turnaround is Phoenix-based First Solar, which says that the technology could eventually be cost competitive with conventional fossil-fuel sources of electricity.

First Solar has racked up a string of large contracts and investments over the past year for its thin-film technology. First Solar closed a $400 million initial public offering in November and clinched a deal three months later to supply a 40-megawatt solar-panel farm in Germany that will be one of the world's largest. And earlier this month, the company revealed that it has signed long-term contracts with European and Canadian buyers to supply 685 megawatts of modules worth $1.28 billion. The latter figure is especially impressive considering that all the solar-module factories in the United States shipped less than 200 megawatts' worth of photovoltaics last year.

Ken Zweibel, who directed the U.S. National Renewable Energy Laboratory's (NREL) Thin Film Partnership Program for more than a decade, says that First Solar has "broken out of the pack" by simultaneously achieving low-cost mass production and respectable power output. Zweibel, who left NREL in January to launch his own thin-film company using similar technology, expects more improvement on both fronts. "Cadmium telluride has a clear route to cost competitiveness with conventional energy," he says.

Thin-film panels are produced by layering shallow coatings of semiconductor materials on sheets of glass, plastic, or metal--a seemingly simple concept that is hard to implement on a large scale. Cadmium-telluride panels in particular seemed finished five years ago when BP Solar, an arm of the London-based oil company, shut down what had been the largest cadmium-telluride commercialization effort.

Story continues below

BP Solar had opened a cadmium-telluride module plant in 1998 designed to make eight megawatts of modules per year, but it never exceeded one megawatt. Creating films to exacting specifications proved harder than expected. And the company was concerned about the product's image, given the use of the toxic heavy metal cadmium. And BP stumbled in the market when the efficiency with which its first commercial cadmium-telluride modules absorbed solar energy slipped from 8 percent efficiency to 6 percent after just a few weeks on rooftops.

First Solar, founded in 1999 from a predecessor startup called Solar Cells and with an infusion of $250 million from Walmart founder John Walton, kept on tweaking its manufacturing process. The company addressed concerns about the toxicity of cadmium by creating a recycling program guaranteed to take back panels at the end of their useful life. Company officials would not comment in advance of an earnings statement. But company documents say that it progressively ratcheted up production at its first plant in Perrysburg, OH, from a few hundred kilowatts of modules per year in the early years to 75 megawatts last year. First Solar now produces more than 100 megawatts' worth of panels per year, thanks to a new plant in Germany.

Comments

  • good - $ / KWH the holy grail
    It's good to see a lower efficiency PV system also working to lower the $ / KWH price.

    As great as the high efficiency triple junction cells are, they're very expensive.  Adding concentration and tracking increases complexity and expense.

    Which will eventually win?
    High cost / high efficiency versus low cost / low efficiency?
    (In the battle for lowest $ / KWH)
    Rate this comment: 12345

    nekote
    07/27/2007
    Posts:138
    Avg Rating:
    3/5
    • Re: good - $ / KWH the holy grail
      It would seem there probably is a place for both ends of the cost/performance spectrum.  In situations where installation labor costs are high, such as retrofitting a house in the U.S. with solar, the high efficiency would probably out weigh higher module costs, since the module cost is much less of the overall costs.  In situations where the installation costs are lower, such as developing countries with cheaper labor or new constructions (especially where the solar might also serve to replace roofing or other structures), it might make economic sense to use cheaper though lower efficiency.
      Rate this comment: 12345

      rhapsodyingl...
      07/31/2007
      Posts:55
      Avg Rating:
      4/5
  • Depreciation
    Peter is obviously not an accountant. Depreciatio may affect income (and taxes) but a depreciated factory is not a liability as he implies. To the contraryy a depreciated plant is typically "written off"and is no longer a charge to current operations and cost
    Rate this comment: 12345

    don.hutchins...
    07/27/2007
    Posts:12
    Avg Rating:
    5/5
    • Re: Depreciation
      I am not an accountant, but that's beside the point. As you have observed, a depreciated factory is no longer counted against costs. In First Solar's case, this should contribute to their current low cost of production. In the next few years as First Solar shifts to producing mostly at new factories, it will lose this financial advantage. That could make it harder than it might appear for the company to achieve the substantial price drop promised for 2012.
      Rate this comment: 12345

      pfairley
      07/27/2007
      Posts:9
      Avg Rating:
      3/5
      • Re: Depreciation
        Actually according to page 39 of the 2006 annual report you are mistaken about the depreciation.  In fact it increased by $5.9 million at the Ohio facility as the result of a doubling of capacity during 2006.  So in fact the recent steadily improving numbers take into account the increase in depreciation and a huge amount of overhead associated with testing new plants as they come on line along with the full costs of operation (equipment, staff, materials) without yet having any saleable production from the new plants.  They should see significantly lower costs as one time startup costs are replaced with normal operational costs.
        Rate this comment: 12345

        WalterEdward...
        07/28/2007
        Posts:1
        • Re: Depreciation
          In other words, capital costs are rising -- just as one would expect. Note that the lowest capital cost estimates for solar manufacturing facilities are in the $1/watt range, suggesting that the $5.9 million charge you cite can not represent the full cost of their corresponding 25-50 megawatt expansion.

          To get back to the larger point: It is not that First Solar can not or will not deliver on its price goal, but rather that one must not go forward under the illusion that meeting such an ambitious goal will be easy.
          Rate this comment: 12345

          pfairley
          07/28/2007
          Posts:9
          Avg Rating:
          3/5
  • Possible for WEDS
    The thin film may well be a good candidate for a described three tier intercontinental Northern Hemisphere tie-in called WEDS - World Energy Distribution System.
    Rate this comment: 12345

    Futurenow
    08/01/2007
    Posts:2
  • CdTe
    It is valuable to see a new, thoughtful article like this on CdTe. It's an important new technology because systems are going in at record low prices - a $4.2/Wp 40-MW First Solar/JUWI Group ground mounted system in Germany; and $5/Wp large roof system by the same team (these systems would be in the 15-20 cents/kWh in US locations with good sunlight). These are about 20% less than any other known installed PV prices - I invite others to comment on any other well-documented low cost PV systems, in case I have missed any (these are published prices on the JUWI website).

    CdTe is also a fair example of a prior success of Federally sponsored research in the US, since CdTe has been developed and nurtured in alliance with Federal DOE program through NREL. How many other TW-potential technologies have emerged commercially from such programs? First Solar has been and still is nurtured by the DOE/NREL PV program (as is PrimeStar Solar).

    Information about CdTe has been rare in the past due to an absence of publicity because of the fear that public misunderstanding about using tiny amounts cadmium would be a problem. In the meantime (25 years!), excellent experts like Paul Moskowitz and Vasilis Fthenakis of Brookhaven National Lab have done the spade work to show that safe manufacturing practices and recycling would close the loop on perceived Cd issues. But for this reason, the world is starved for information about this very important advance in practical, low-cost PV.

    Finally, the path forward to low cost comes from both technical advances (improved efficiency, optimized processes, thinner layers with less material cost...) and volume production (economies of scale). Although it is hardly guaranteed, it is logical and clear all the way down to about 5 cents/kWh (about $1.2/Wp installed) for large systems (over 1 MWp).

    The progress of solar technologies, on which so much depends in terms of climate change, energy supply and diversity, and world sustainability (economically and politically) will only grow stronger now that the power of investment is beginning to be substantial. In comparison, prior progress will seem inconsequential and idiosyncratic when we look back on it. But sustained development in solar requires the period of transition to cost-competitiveness best exemplified by the current US bills to support commercial solar with tax credits, renewable energy portfolio standards, and other programs like the excellent European "feed in tariffs" (which pay per kWh and thus are more aggressively cost-oriented).
    Rate this comment: 12345

    kzweibel
    08/05/2007
    Posts:1
    Avg Rating:
    4/5

Log In

Forgot your password?     Register »
Advertisement

Videos

Laser-Triggered Chemical Reactions
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 © 2009 Technology Review. All Rights Reserved.