Coyote Blog

We Phoenicians, who live in one of the best solar sites in the world, have been anxiously awaiting a solar electric technology that makes economic sense.    I have a couple thousand square feet of nice, flat room that is just begging to get be off the grid.  Already, solar is economic for individuals in Phoenix, but only if you are willing to soak American taxpayers and your neighbors for 85% of the costs.  It would be nice if it were, you know, actually economic and not just subsidy bait for tens of thousands of dollars.  I have dug into many analyses that claim that solar has a 5-7 year payback, but never seen one that achieved these returns without substantial subsidies and rebates (beware the term "energy payback" which is not the same thing as investment payback (pdf))

For a while I have said that I thought traditional silicon/germanium IC-like wafer processes for making solar cells was just never going to get there, and that some other technology was necessary.   This might be one such example:

JA Solar, one of the big players in the solar industry, is working with Innovalight to commercialize the latter's method for making silicon-ink-based, high-efficiency solar cells, the companies said this week.

... The solar cells are created by pouring an ink solution incorporated with silicon nanoparticles and then decanting the excess liquid to leave behind a crystalline silicon structure.

At the time of the 2007 announcement, Sunnyvale, Calif.-based Innovalight claimed its method not only resulted in solar cells that were cheaper to produce by as much as half, but that the crystalline structure resulting from the process made its cells more efficient at converting electricity.

Those claims now appear to be validated.

On Tuesday, Innovalight announced that an independent study of its method by the U.S. Department of Energy's National Renewable Energy Laboratory and the Fraunhofer Institute for Solar Energy Systems in Germany confirmed that its silicon ink-based cells "demonstrated a record 18 percent conversion of efficiency."

The 18% conversion efficiency is close to a record for thin films, but must be the "record" for production models, since higher conversions have been achieved in the lab.  18% is very good for a production device, particularly if it is cheaper to manufacture than current cells.

All those confident in our ability to ramp up things like wind and solar quickly should take a long look at T. Boone Pickens decision to virtually abandon billions of investment in wind.

One of the ways I think our potential to increase renewables is over-stated is that the government has begun lumping hydro power into wind, as in these charts.  They show "renewables" as about 9% of electricity production.   Increasing this to, say, 20% seems daunting but doable - after all, we are just doubling it.

But in fact, almost all of the 9% is hydro power, and that is not going to increase (in fact environmental presure is actually to destroy several hyrdo facilities to allow the rivers to run free).  This means that to get total renewables to 20%, other renewables like wind and solar will have to increase from about 1% to 12%, or a twelve fold increase.  This is much more daunting, especially since a raft of subsidies and incentives and programs have gotten us to just 1%.

Postscript: Owning a home in Phoenix with a big flat roof, there is no one in the world rooting for solar to be economic more than I am.  I have run the numbers recently, and taking advantage of all government subsidies, the investment has about an 8-10 year payback.  It's just not there yet.  Further, I worry that the current silicon/germanium IC technologies are dated and dead end.  I fear that buying solar now is like buying the last IBM mainframe before PCs came out.  I have a ton of confidence in the innovativeness of man, and believe that a real solar breakthrough will occur in the next 10 years.  Wind, on the other hand, is never going to work.  It is the ethanol of electricity production.