Implications:Until fairly recently the source of solar grade silicon for manufacturers of photovoltaic solar cells was excess polycrystalline silicon from processes intended to produce ultra pure silicon for the production of electronic integrated circuits (chips). This was because the processes to purify and produce silicon  as ultrahigh purity (99.9999+) single crystal 'ingots,' for slicing into 'wafers' themselves to be the base upon which integrated circuit chips were formed and then removed by slicing and dicing, were complex and slow and yields were never 100%. The waste, i.e. either polycrystalline material or that of less than ultra high purity, was sold as 'solar grade.' Lately a number of makers of ferrosilicon have announced that they have solved the problem of turning their 99.7% metallurgical silicon directly into 'solar grade' silicon by the ton.  If this is true then silicon PV solar cells may become competitive with thin film cadmium telluride PV solar cells.

Analysis: 
The 'independent' solar cell manufacturing auditing firm, Photon, in a recent audit of the Canadian metallurgical firm, Timminco, to look at a process developed by a Timminco subsidiary, Becancour, for mass producing what is known as UMGSI, upgraded metallurgical grade silicon, which means solar grade silicon produced directly from metallurgical grade silicon, stated that the global demand for PV solar cells in 2010 will leap from 2007's 3-6 gigawatts to 26 gigawatts.

Photon predicted that the price of the solar grade silicon would not be as much of a factor as the availability of the required amount in 2010. Perhaps more than 50,000 metric tons.

Photon repeated Timminco's claim that it would be able to produce 14,400 mt of UMGSI in 2010 if its funding and equipment programs were on track and on time, but that, even so, Timminco would only then be the fourth largest producer of UMGSI at the end of 2010.

There is no commonly used process for producing UMGSI, at least not one that is commonly acknowledged. The large producers whose ranks Timminco hopes to join each guard their proprietary process and make announcements about the processes' efficiency, its ability to produce large tonnages, and their contracts with key customers on a regular basis.

A great many market insiders are skeptical, because they see a bubble already forming in alternative energy stocks, such as solar energy conversion related ones, and they believe that a lot of the secret processes are intended to promote the stock rather than increase the availability and decrease the cost of producing UMGSI. It is instructive to note that Timminco has gone from $0.30 to $30.00 a share in just a couple of years as an example of a skepticism inducing situation.

So, how does the potential for mass producing UMGSI at lower prices than today affect the non silicon thin film PV solar cell industry?

Probably the largest maker of such non silicon films is First Solar, which concentrates on thin films of cadmium telluride as its PV material. It has been stated that one gigawatt of such thin film cells requires 30 mt of tellurium.

The problem is that this means that 26 gigawatts would require 780 mt of tellurium. There is probably enough tellurium production potential to meet this demand, but with today's estimated production of tellurium at less, perhaps much less, than 500 tons and with no visible surplus from tellurium's major demand as an alloying element in specialty steels and copper it is clear that there is no possibility of the demand for enough tellurium to produce 26 gigawatts of electricity in 2010 being met. It looks like any increase in the use of tellurium for thin film PV solar cells will have to be met by new production and that the bulk of the projected 26 gigawatts of PV solar cells built in 2010 will be made from solar grade silicon or not made at all.

The race to produce very large new tonnages of solar grade silicon is under way. There is no hope of doing this economically by building plants to manufacture ultra pure electronic grade silicon and operating them in such a way, inefficiently, as to force the production of lesser quality solar grade material.

The answer is to hope that UMGSI processes are already successful, and the only way to test this, without revealing secrets or destroying competitive advantage is for the various companies that are announcing such processes to allow their end customers to reveal what quality and in what quantity they, the end users, are now receiving UMGSI.

If no such information is forthcoming then the bubble may burst.

(Article from Gerson Lehrman Group)