[Wien] Re:How to calculate the reflectivity when calculating optical properties

[Fred Nastos]

> Do you mean I should use the imaginary and real part values
> in the file case.epsilon to calculate the reflectivity?

I mean only what is written at the top of page 131 of the
usersguide.  That is, "All dielectric tensor components can
be found in file case.epsilon[up|dn]."

> Today I tried to calculate the reflectivity of anatase TiO2, but
> the obtained values of reflectivity are less than 35% in the
> range of 200-900nm. While my experimental values are
> more than 80%. What's the matter?

The reflectivity comes after a lot of post-processing.
Why not go back to the more "raw" theoretical results.

I suggest you do the following:
1) find previous DFT-based calculations on anatase TiO2.
One that seems particulary relevant is PRB 51, 13023 (1995).
It even mentions reflectivity calculations.
2) Compare your band structure, DOS and computed dieletric
functions to the ones in the above paper.
3) Check your scissor-shift.  Did you use a scissor shift?  Note
that (for normal incidence) the reflectivity is
	r = (n-1)^2/(n+1)^2.
A scissor shift lowers the value of epsilon(0), which lowers
the theoretical value of the reflectivity.

Good luck.