Many posters on and off board are asking if the world market of Gallium Arsenide can supprot a massive shift from Si to GaAs.

Please have a look at recent publications of the US Geological Survey - I guess these figures are quite accurate:

Gallium: http://minerals.usgs.gov/minerals/pubs/commodity/gallium/mcs-2015-galli.pdf

Arsenic: http://minerals.usgs.gov/minerals/pubs/commodity/arsenic/mcs-2015-arsen.pdf

Possibly it is worth a read, as you can learn a lot about resources, applications and material streams.

I am resuming:

World production 2014:
Ga: 440 MT (Metric tons)
AS: 46,000 MT

So it is evident that Gallium is the limiting factor.

World primary gallium production capacity in 2014 was estimated to be 680 metric tons; refinery capacity, 230 metric tons; and secondary capacity, 200 metric tons.

This means the production capacity is now about 2.5 times the actual production of Ga. A lot of potential to grow!

We should not forget that for POET, we need highest purity Ga and As, so this purification step could be a limiting short term factor. From the report we can learn that 2014 the US produced about 65 MT of high purity GaAs. At the moment main applications are RF (analog IC) and white LED, which for long will be the major factor of demand, as here the market still has immense potential.

Can someone of our experts estimate how much GaAs is needed to produce an "average" 40nm chip (and how much will finally be fixed in the chip - as most of the rest will be recycled...)?

So my 2 cents: I do not see a supply problem. An additional factor: If there would be any proble with supply, not all Si need to be replaced by POET. There will be a plenty of applications left for Si.

An interesting read at the USGS is also: "Gallium, a smart metal": http://pubs.usgs.gov/fs/2013/3006/pdf/fs2013-3006.pdf

Possibly worth to put on our link list?