Mike, I still think that the major stumbling block is the time that would be required to take your test.
You could reduce both setup and trial time as well as material needed by testing for 1 sample in 5 containers.
Admittedly, you'd have to get 10 from 10 for a 10 pass trial to beat 1:1Million odds, but has JREF determined that these MUST be the odds you need to beat for the preliminary?
If not, 10 passes of 5 containers (i.e. 50 in total) would require a success rate of 8 to beat odds of 1:10,000. Perhaps sufficient for a preliminary?
ETA. I see from post 4 from JREF in the original Challenge thread that the odds given to beat for edge are indeed 1:10,000.
"A positive result for the preliminary test will be identifying the target substance (of either gold or silver) correctly in seven out of the ten sets of ten."
Source for my "calculation" of odds.
I don't think they would go for 50 but yes that would be a one day trial.