[Wien] calculations of spin-dipole hyperfine fields

[Igor Mazin]

Stefaan Cottenier Stefaan.Cottenier at fys.kuleuven.be
wrote:

> You need to run only lapwdm -up. The last column of
case.scfdmup then gives you the field you want (in
Tesla).

Thanks Stefaan, but this is true only for lapwdm -so
-up. At least, in the version I am using (2k_03).
lapwdm -up correctly computes the spin-up part, and
lapwdm -dn the spin-down one, and I have to substract
them by hand.

>For orientating the hff, wouldn't it be better to do
it by spin-orbit coupling (the angles in case.inso)
instead of by an external field? 

I needed non-so calculations to compare with another
code. But I guess I can use Peter's recipe and run
lapwso with 0 orbitals. There is, of course an awkward
option of rotating the lattice, but I was looking for
a more elegant solution. And, of course, I need an
external field to have a non-zero hyperfind field
anyway :)

Peter Blaha pblaha at theochem.tuwien.ac.at wrote:

> When using a magnetic field (along a certain
direction), you MUST use lapwso; i.e. you must use  
runsp -so -orb

I see. The output format is somewhat confusing, it
looks like it explicitely calculates the averegae of
r^2-3z^2 no matter what. I will try it with SO right
now, thanks!

> If you like, you can switch off all SO interactions
in case.inso

Indeed! Thanks again.

Igor

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com