XRF through a slab?

I tried to XRF an archaic Greek coin in an NGC slab.

I got very strange results: LE 97.82%; Ag 2.15% (Silver); Au 0.024% (Gold); Pb 0.010% (Lead)

LE stands for “Lighter Elements”. For this machine, this means Mg, Al, Si, P, and S. (The machine has a way to break this down, but it isn’t my machine, and I didn’t figure out how in the time that I had.)

I gave up after a single test.

According to the blog from a different XRF manufacturer, XRF can be used through slabs. So I decided to take a look at the numbers I got.

If I throw away the LE, and normalize the silver, gold, and lead so they add up to 100%, I get:

       2.15
------------------ = 98.4% silver
(2.15+0.024+0.010)

Using similar calculations, gold is 1.1% and lead 0.5%. Those figures are about right for the 3.9g series of “Parion” (actually Thracian) coins.

Perhaps it is possible, with machine settings or calculations like the above, to perform XRF through slabs. Has anyone tried it before and after cracking slabs? Is there a technique that works?