Last Friday I attended a very informative solar technology workshop put on by the IEEE at Freescale in Austin. It brought together experts in both microelectronics and solar technology who provided a wealth of information about both fields and the intersection between them.
Status of Photovoltaic Technology
First up was Dr. Jeffrey Mazer, an engineer with the U.S. Dept. of Energy’s Photovoltaics Program, who spoke on the “Status of Photovoltaic Technology and the DOE / EERE Photovoltaics Program.”
Photovoltaic (PV) solar panels are in huge demand. The market grew 51% in 2007 and 81% in 2008. Seven gigawatts of PV panels were produced in 2008, mostly in India and China. The U.S., which in 1995 had 45% of that market, is now down to 8%. As Tom Friedman remarked recently, “So, if you like importing oil from Saudi Arabia, you’re going to love importing solar panels from China.”
Direct PV module manufacturing costs are now down to under $2/watt, which makes them increasingly attractive; the increasing use of cadmium telluride (CdTe) can drive that down to ~$1/watt. In 2008 about 86% of PV modules were made from crystalline-Si, thought the use of thin-film PV is growing fast due to it much lower manufacturing cost. More than half of U.S. production in 2008 was thin films. Thin films now make up 13-14% of the PV market worldwide and over half in the U.S.
You Got a Problem With That?
Photovoltaics aren’t without their problems. Dust builds up on the modules and obscures sunlight, creating a need for anti-soiling coatings. Inverters and chargers need to be more reliable–they currently suffer from thermal problems.
There are also problems with the basic physics. A solar cell must be able to absorb photons; separate the excess (photogenerated) carriers; and transport them to external terminals. This leads to two fundamental and several difficult challenges.
Fundamental problem #1 is the rapid cooling or thermalization of hot carriers. Energetic photons create carriers too far into the conductance band, where they lose kinetic energy through electron (hole) –phonon scattering mechanisms. The use of multijunctions can help conduct away excess heat, but the process is expensive.
Fundamental problem #2 is the inability to absorb low-energy (sub-bandgap) photons, which further limits their efficiency. For example, for crystalline-Si, infrared (> 1150 nm) is not absorbed.
Mazer’s suggested solution for addressing both problems is quantum dots, which slow down cooling dynamics, allowing impact ionization (inverse Auger process) to produce several e-h pairs from one photon. Quantum dots can be induced by creating a superlattice. For barriers < 4 nm, the wave functions overlap, and minibands form. Minibands might serve as electrodes for collection of hot carriers before they can cool, or absorb sub-bandgap photons in a p-i-n cell.
Difficult though not fundamental problems include high materials costs due to the large size of PV cells (–>use thinner wafers); balance of systems (–>more reliable inverters); and various issues with glass. Mazer predicts that solar will consume 5% of worldwide glass output by 2015-20; however, the low profit margins on such glass creates little motivation for new low-Fe glass plants. So expect to see a possible shortage of solar glass in 2015-2020 that will parallel the shortage of silicon when PV panels really took off a few years ago.
Mazer spent some time explaining different architectures for both silicon and thin film PV cells. The real imperative for the technology is to increase efficiency while bringing down costs. The bottom line, in short, is economic. Currently the cost for producing electricity from solar cells is about $.17/kWh; according to Mazer, that needs to come down to $.05-.10/kWh for solar to be competitive without requiring subsidies. We’re getting there, but we’re not there yet.
Chew On That For a While
On a closing note I had lunch at Dr. Mazer’s table where he fielded questions from fellow hungry engineers. One said he hoped solar could “help relieve our dependence on foreign oil.” Mazer pointed out that only 3-4% of U.S. electrical power is generated by petroleum, so that wasn’t in the cards. Considering that most U.S. electrical power comes from coal-fired and nuclear power plants, hopefully solar will reduce our dependence on those sources.
Reduce our carbon footprint and ‘uranium footprint’ at the same time. I could get behind that.