This week’s Economist has an article about solar power, focusing on a recent conference in San Francisco. As the article points out, one of the main problems with solar cells is that they are not very efficient. Right now they are quite expensive to produce, but if we could double the efficiency with which they turn sunlight into electricity then the effective price is halved (assuming no increase in the cost of manufacture). Therefore prospective efficiency gains of the type described by the article are well worth looking at.

But there are other options for making solar power cheaper. The easy one is actually simple economies of scale. If there world demand for solar cells were greater then individual cells would probably become cheaper to produce. Possibly much cheaper. I mean, look what happened with memory chips. Unfortunately this is a chicken and egg problem, and one that can probably only be solved by massive government subsidies.

Another option, which I spent some time discussing with Risto Isomaki at Finncon, is concentration. If you can pour more sunlight onto the solar cell then you get more electricity even though the efficiency is low. A definite advantage here is that you only need mirrors to concentrate light, and mirrors are cheap compared to solar cells. You also need a lot less in the way of heavy metals to make a mirror than to make a solar cell, so the manufacturing process is much cleaner. (The usual environmentalist complaint about solar power is that making the cells is a heavily polluting activity.) The problem with concentration is that it heats up the cell, and the performance of solar cells drop off it they run too hot. Research is therefore concentrated on either making cells that can run at higher temperatures, or at cooling the cells when in operation. Anyone interesting in learning more should try this article.