As mentioned in some of the comments to the posted article, I want to know if getting some spectroscopy on those white spots couldn't be possible. Perhaps summed over a quite long series? (Wouldn't it be a thrill if they were metallic?)
Just subtract the known spectrum curve of the sun, and check the difference.
I would guess that it's a very reflective crystalline structure. You can see that as the bright structure goes beyond the terminator, it is still visible for a frame, but it completely disappears the frame after. It's hard to say either way though as we don't know the specs of the camera. Lots of materials reflect light differently at wavelengths outside our normal vision.