[Wien] Unit of :ORBxx

[Stefaan Cottenier]

>I also think this is the case. As a little test, I try to calculate :ORB
>of a free Fe atom by constructing a fcc Fe with a lattice constant of 8A
>(3 times the real fcc Fe bulk). I turn on the orbital potential first, and then
>add spin-orbit. Both OP and SO have their M's pointing in (0 0 1). Then
>I get :ORB to be 0.2. How do I relate this number to the fact that a free Fe
>atom has L=2? I thought I would have obtained :ORB very close to 2, though.
>  
>
I never did that particular test, but I would indeed expect something 
much closer to 2. Some hints:

* Your lattice constant is a bit small. The FAQ recommends 30 bohr for 
free atom calculations ( http://www.wien2k.at/reg_user/faq/cohesive.html )

* The orb package calculates properties only inside the muffin tin 
sphere. Try to take the largest Fe sphere that still makes sense (say 
2.5 bohr), and see whether you get closer to the desired value.

Stefaan

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm