"Messy" Reflectivities

Detlef Smilgies

Thicker organic films, i.e. between 100 nm and 1000 nm, are often very rough. Can we still learn something about such films when there are no mor Kiessig fringes, be it that they are damped out by the Roughness, be it that they are smeared out by the finite resolution? The answer is yes, if we look carefully at the critical angles. If we crank up the roughness, the following will happen:

first the Kiessig fringes disappear
then the resonances between the critical angles will disappear

When we still have the wave guide resonances we proceed as follows

we determine the critical angle of the film. From this we get the average film density, if the chemical composition of the film is known.
we vary the thickness until the number and location of the resonances is reproduced
we fit the roughness for overall good agreement

Step 2 and 3 may have to be iterated.

Can we still learn something about the film when we do not see the resonances any more? Yes, is the answer, if we proceed this way:

again we fit the film critical angle and get an estimate for the density
if the stoechiometry is known, we get both optical parameters δ and β
we fit the thicknes to reproduce the attenuation of the reflectivity at the critical angle of the substrate
we fit the rounding off at the film critical angle and the initial dip to get the roughness

Now, we have to be a bit careful not to leave the validity of the NCVV roughness model - as long as s is small than 10% of the thickness we should be fine. For a 3000 Å film this actually means that we can model huge roughnesses of up to 300 Å!