CapMan...
wasn't parts per trillion for the coating?
By using organic TMA, it doesn't bring metals (and it says so in the patent). There's also no reason to think the chemical is not recycled. The only limitation on how dirty the constituents are, is how fast the filtration fails. You end up with a tradeoff between the quality of feedstock and filter replacement/backwashing. That's where the real expense is. Downtime and hardware.
8.5 pH is not going to take much TMA either and since that is reused, they seem to hint that pH control is via HNO3 injection which (surprise) getters metals nicely and actually renews the TMA. It precipitates from the barium nitrate impurities AND any metals it might bring with it.
Besides, stage three is where the real meat is here and you know it. Anyone else find it ironic that kerosene is used to manufacture a device that will help substantially reduce petroleum consumption? Third stage processes are also using recycled chemicals.
Zoom out from oxidation state numbers, and material compatibilities for a minute and consider this for what it is. The impurities are concentrated in solids and filtered out. In terms of chemical use, this is a very mass efficient manner in which to produce a chemical of high purity. I think the most expensive part of this process is going to be the filters and the downtime for replacement and periodic backwashing.
I have 15 years direct experience in semiconductor wet chemistry equipment and process. The product for my experience is a solid instead of liquid so I understand the major differences in terms of material handling. I know there are limitations to my understanding but I also see how this is unique and non-obvious.
“Historically, the claim of consensus has been the first refuge of scoundrels; it is a way to avoid debate by claiming that the matter is already settled.”- Michael Crichton

