Photovoltaics Might Be Silicon But The Technology Is Not Following Moore’s Law

Photovoltaic specialists met last week, May 12-16, in San Diego under the auspices of the IEEE Electron Devices Society, for their 33rd annual meeting. For the first time the meeting included a two-day breakout session, The PV Accelerator Forum, devoted to exploring how photovoltaics can be kick-started to achieve an earlier commercial breakthrough. There were some substantial surprises.

If you asked a solar expert ten or fifteen years ago what the game plan was for photovoltaics, the gist would have been this: develop silicon cells, relying on scraps and techniques from the semiconductor industry, without expectation of a commercial breakthrough; then turn to second-generation thin-film materials like CIGS and cad-tel, which would be much cheaper and more fit for mass production. By early this decade, however, it seemed clear that PV was not shaping up as planned. The second generation materials were not materializing on schedule, and the cost of solar electricity was still nowhere near competitive. Particularly disconcerting was the 2002 decision of British Petroleum, which was billing itself as the world’s biggest solar company (among other things), to terminate U.S. production of cad-tel and amorphous silicon cells, as reported in the January 2003 issue of Spectrum magazine.

Now there are some new twists and turnsâ¿¿essentially, three very positive developments that would not have been generally anticipated a decade ago. First, silicon-based solar technology has decoupled from the semiconductor industry and is achieving steady cost reductions, so that those following PV discern a kind of Mooreâ¿¿s law at work. In 2005, production of silicon for solar cells already surpassed production of silicon for semiconductors.

Second, the industry has become so confident in that evolutionary path, policymakers and planners have started to set dates when they expect PV-generated electricity to be competitive with the major sources of electricity sold on the grid now. And third, while the incremental path promises a commercial breakthrough within ten years, it’s suddenly looking like second generation technology may be arriving after allâ¿¿in which case wide commercialization of PV could occur much sooner.

In recent years, global PV production has been increasing at a rate of 50 percent per year, so that accumulated global capacity doubles about every 18 months. The PV Moore’s law states that with every doubling of capacity, PV costs come down by 20 percent. In 2004, installing PV cost about $7 per watt, compared to $1/W for wind, which at that time was beginning to stand on its own feet commercially, Last, year, as recently noted in this blog, average global solar costs had come down to between $4 and $5 per watt, right in line with the PV Moore’s law. Extrapolate those gains out six or seven years, and PV costs will be below $2/W, making photovolatics competitive with 2004 wind.

Remember, wind electricity generally is generated in large farms, so that its price has to be competitive with electricity generated from other sources that’s the wholesale electricity cost that accounts for only about half of total electricity costs in a typical customer’s bill. But solar, being distributed, competes with the retail price if the PV generating cost is comparable to the total delivered cost of electricity, which can be as high as 20 cents per kilowatthour in the United States and upwards of 30 cents in Japan, thatâ¿¿s good enough.

Planners and regulator are starting to believe in the PV Moore’s law. The European Union’s PV Tech Platform has set the year 2015 for achieving grid parity the point where solar electricity can be sold competitively into the grid. As early as 2010, solar electricity prices in extreme southern Europe might go as low as 17 or 18 cents/kWh. California also expects to see grid parity within a decade, and Southern California Edison has a program to put subsidized PV roofs on large commercial buildings, predicated on the goal of obtaining PV capacity at a cost of $3.50/W within five years.

So some noteworthy things have happened on the way to this year’s PV accelerator forum. But what was getting the most buzz in the technical conference, which attracted a record number of attendees from around the world, was next-generation PV. Sessions dedicated to next-generation materials like cooper indium diselenide and cadmium telluride were packed to the gills, with people craning their heads in from the hallways to catch snatches of talks. One company is particular has been growing like gangbusters in the last couple of years, with a rather simple CdTe module that it claims to be producing at a cost of barely over $1/W.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s