Silica to Silicon

In a recent post I made an aside about the difficulty of making semiconductor-grade silicon from sand. Such silicon is used for electronic chips, and also for solar panels.

The information I could find on the internet was not particularly clear on the issue of raw materials for making this high-purity silicon. Science News has a 2005 article (Sun and Sand: Dirty silicon could supply solar power) that implies the problem exists by describing a possible new method for solving it.

My partner Wade helped me out with the following explanation. The more junk in the raw materials, the more work is needed to purify it. This is not rocket science but takes lots of work and lots of capital. The primary issues have been (1) allocation of capital in the presence of demand risk, and (2) plant construction lead time (compounding #1). There are new technologies for turning sand into silicon, but the real value seems to be in the project finance.

The last note from Wade is telling. We at Sigma tend to avoid the kind of project where the financial engineering is more important than the real engineering. So, much as I would love to see Israel's silica valley become a real silicon valley, perhaps using all that solar energy to make it happen, I will have to watch from the side, if it happens at all.