1) The amount of gypsum available is not quantified properly, in the context of what was in the towers and might have been present. I see no reason to expect the small amounts Cole uses to be meaningful.
On top of that, there's an argument to be made that as far as drywall being the source goes, there should be a large excess of it. Drywall is indeed cited as the most probable source of the sulfur, but as some here, including
Ryan Mackey, pointed out: Drywall doesn't release its sulfur easily. So on a per-volume or mass basis, you'd need a huge fire and a large amount of drywall relative to the amount of steel you wanted to affect. And recall: Relative to the total amount of steel at Ground Zero, it was a small amount of steel that ended up showing signs of such corrosion. The pieces the Worchester folks studied included a perimeter column, a few pieces of paneling, a wide-flange column's end they think came from building 7... maybe some other odds and ends, but it wasn't much, I think. NIST wrote material on 2 columns (one of which, K-16, was the same one looked at by the WPI group), three panel seats, some number of web plates, and a couple of pieces of "floor truss material". More may have been affected, but that's all that was identified.
Anyway, as Frank Greening pointed out in his sulfur paper:
"In the present context it is not simply the presence, but the mobility, of sulfur in the WTC that is of interest."
... and earlier in the thread, he was quoted as hypothesizing a source of Cl - such as PVC - as being necessary to help liberate the sulfur. Which is logical, and a compelling argument. At any rate, it's not only true that the amount of gypsum was poorly quantified, but on top of that, there's a misunderstanding of the amount of gypsum that would be needed relative to the amount of steel that's to be corroded. As well as a failure to account for the difficulty in liberating the gypsum (and therefore the sulfur component) from it. That all ties in together as a critique of the experiment.