[Wien] [Wien2k Users] Spin orbit coupling of Atom with many nonequivalent positions

[Stefaan Cottenier]

> I did it for all allotropes for uranium (with lo+apw for all orbitals)
> Keeping Emax as 2.50 (as per your suggestions in the book you have posted in
> Wien2k website)

That was for a different purpose (plotting the DOS for higher eigenvalues).

> If Emax is changed to 5.0 for example (I ask this question from the mail on
> hcp Tb; the case.in1 had 5.0 as Emax), then should it really be done (for my
> case?). 

The influence of spin-orbit coupling is calculated by a second 
variational step, using a different basis. Emax plays the role of RKmax 
for that second basis. You have to test total energy (and perhaps other 
properties that are affected by spin-orbit coupling, such as the orbital 
moment) as a function of Emax (start at 2.0, go in steps of 0.5, it is 
unlikely you will need to go beyond 5.0).

Stefaan