I'm using a germanium cell to confine a liquid crystal monomer and study it with FTIR in transmission geometry.
I have use previously CaF2 and ZnSe substrates sputtered with gold obtaining a simple interference pattern of the cell gap (although very low transmission).
When making a cell with germanium, i get 2 interference patterns, one agrees with the 10µm silica spacers used for the cell construction, but i get a second pattern for about 150µm (with n = 1).
Substrates alone don't show interference (so is not between 2 faces of germanium and even if it was, the thickness wouldn't match even using Ge refractive index).