[Wien] Choice of magnetization axis in SO calculation

[Marcos Veríssimo Alves]

Hi all,

I have searched the list and the literature proposed in the Wien2K site, and
I haven't found an explanation to my doubt, so I guess it'd be appropriate
to ask it in the list. It's a very basic question, but it didn't occur to me
in the last Wien2K workshop in Nantes...

I have a system which is spin-polarized. I am about to include spin-orbit
coupling in my calculation, but I am unsure of how to chose the
magnetization axis for initso_lapw. From the user's guide, it states that
the axis of quantization can change the symmetry of the system, which I can
understand. However, looking at P. Novak's notes on SOC for Wien2K, I see
that in the example of Ni, he chooses two different directions for the
magnetization axis, which give different results. Nevertheless, it is not
stated which of the two choices is the correct one, if there is such a thing
after all.

The thing is, I have a system which has a funny feature at the DOS, and I
would like to check the robustness of these features against SOC
(essentially, see if SOC induces mixing of different m for l=2). However,
since different quantization axes yield different results, its correct  (or
a reasonable one, at least) choice turns out to be crucial. The problem is,
I have no clue as to how to choose the quantization axis.

Should the quantization axis be chosen parallel to the direction of the
magnetization in the no-SOC spin-polarized calculation? If so, how can I
check for the direction of the magnetization in my system from the Wien2K
output? Should any other criterion be used for the choice of the SOC
quantization axis?

Best regards,

Marcos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20100806/220c7ed0/attachment.htm>