Brian,

Thanks for your reply. When I posted the news item excerpt, it wasn't so
much to say that I regarded the predictions as realistic as it was just
to draw attention to an intriguing claim that if true would be hugely
important. But your question prompted me to do some research. 

The reporter, who seemed to be an expert in this field, evidently did
not think to inquire how much of the enabling material would be needed
to make a dent in the world's energy production. (His reference to
polysilicon was merely to acknowledge that there happens to be a crunch
right now for this material, in a grade that can be used for producing
photovoltaics, until more production facilities can be built - not that
silicon itself is in short supply.) Here is some information that may be
helpful.

I had imagined that that the material the photovoltaic companies are
counting on was some organic semiconductor, but have now confirmed your
statement that it is copper indium gallium diselenide. Three out of the
four elements involved in this compound are available in only limited
quantities, so I can understand your skepticism. 

On the other hand, I would think that the people running (and investing
in) these companies would have considered the present and future price
and availability of their raw materials as a part of their business
plan. According to the news item, the predicted cost to build their
product is one-tenth of the cost of present photovoltaics. If that
should prove true and can be sustained, it qualifies for the term
"disruptive technology." The fact that there are several such companies
suggests that they think they have a future. So what are the facts?

The abundance of elements in the earth's crust can be seen graphically
in Figure 4 of http://pubs.usgs.gov/fs/2002/fs087-02/ and numerically at
http://tinyurl.com/ykaqg7. From the latter source, the abundances (in
atoms per billion atoms of rock) are shown below (concentration in ore
is of course much higher).

Cu  22000

Ga  5500

In  30

Ag  20

Se  10

For perspective I have included in this list Ag (silver). Last year over
600 million troy ounces of silver were mined and refined. If similar
quantities of indium, gallium, and selenium were produced, I think that
we could make quite a bit of copper indium gallium diselenide. That
semiconducting material is present in the photovoltaic device as a thin
film, presumably vapor-deposited to maybe a few tens of micrometers
thickness, so although I haven't estimated the mass or cost per square
meter (maybe you'd care to), I would think that a little would go a long
way. 

And of course, once the solar cells are fabricated, they are able to
produce electricity for a long time. Unlike an application like silver
in photographic film (which despite its rarity people have been willing
to pay for), it doesn't get used up with use. So the longer the film
produces energy, the lower the cost of the raw materials per kilowatt
hour. That's what makes photovoltaics so attractive. That, and the fact
that it doesn't produce waste or kill birds.

As for the toxicity of Se, I don't see why that should be a greater
concern than for arsenic. I have wondered out loud before why the EU
didn't include arsenic in its RoHS list of prohibited materials. I
suspect that they knew that to prohibit arsenic would be to prohibit
cell phones, and that even they didn't have that kind of power. For
whatever reason, they also did not prohibit selenium. If this technology
were to take off, do you think that they might?

On the basis of the news report I referred to we won't have to wait very
long to assess how well-founded your skepticism is. I'll go on record as
being cautiously optimistic, but as always, I welcome your response.

Gordon Davy