with Mathematica software, I derived the Image Unit position correction formula:
SampleThickness*(n-1)/n
I used the ABCD matrixes of the absorption bench system and the equations for the q parameter.
I cannot attach a .nb file, so I attach the pdf of it