[Wien] Orientation of Atomic orbital Moments

[Hua Wu]

On Thursday 03 November 2005 10:47, Robert Laskowski wrote:
> > If I set the easy magnetization direction along c-axis
> > in the calculations with spin-orbit coupling, but I get
> > the orbital moment, e.g., (0, 0.5, -0.5) uB. That is,
> > besides the c-component, the b-component is also
> > present. Does it mean the (01-1) instead (001) is
> > most probably an easy  magnetization direction ?
>
> it means that orbital moment is not colinear with spin moment.
> In a case of orbital moment whole density matrix is
> calculated by lapwdm, and all components of the vector are in
> output.
>
Does it mean in my system, the orbital moment is 0.7 uB
along (01-1) direction ?

> I think you have to add spin and orbital moments. 

sorry, I cannot fully understand this comment.

> But S_x and  S_y are not calculated, and you can only 
> hope they are zero. 

Since in my system, the S_z is almost saturated, being 
close to the expected maximum S,  I  would assume
the S_x/S_y to be small.  If this is the case, does it mean
in my system, the easy magnetization axis is along
the (01-1) direction, which is also the spin orientation
due to spin-orbit coupling.

best regards -- Hua Wu