[Wien] A new question about the spin texture calculations in WIEN2k

[Peter Blaha]

At least question 2) is easy:

When you run

x lapw1 -band -up/dn; x lapwso -up

then you MUST run:   x lapw2 -band -so -up

Without -band, lapw2 expects the k-points on a tetrahedral mesh and 
stops with a clear error message, which you seem to have overlooked.

After a calculation, I suggest to do:  ls -alsrt

This shows you the last modifications of all files and their length 
(non-zero).

abour point 1) I don't know exactly what you are doing ....

Am 14.06.2020 um 00:39 schrieb Artem Tarasov:
> Thank you for your clarification.
> 
> I have chosen a simple way and made only minor changes in lapwdm.f. Now 
> I save only one value of the weigh(1,:) array, which is read from 
> case.weightup, and put all other values to zero, and do so for all 
> eigenvalues that belongs to the corresponding k-point (it was added as 
> additional regime of lapwdm).
> 
> However, I have noticed two interesting facts:
> 
> 1) Putting all except for the one values of the weigh(1,:) array  to 
> zero always leads to zero spin projections. Nevertheless, if I change 
> these weight zero values to very small ones (0.000001), I get spin 
> projections that are not zero and really depend on energy eigenvalues.
> 
> 2) Running “x lapw2 -so -up” before “x lapwdm -up -so” always leads to 
> zero spin projections. However, I think that everything is ok, if I use 
> “x lapw2 -so -up -band”. I’m not sure that it is right step. I just 
> don’t fully understand a difference between these two options for only 
> one k-point.
> 
> So, I’m confused again but now I feel that I am very close to resolving 
> my long-standing task.
> Would you be so kind to help me to figure out these two facts?
> 
> Sincerely yours,
> Artem Tarasov
> Department of Solid State Electronics
> SPSU.

-- 
Peter Blaha
Inst.Materials Chemistry
TU Vienna
Getreidemarkt 9
A-1060 Vienna
Austria
+43-1-5880115671