[Wien] Clarification on Magnetic Anisotropy Energy calculations

[Peter Blaha]

No, both strategies have in general 2  major flaws.

init_so determines the symmetry (using symmetso) for a specific 
magnetization direction. Obviously, in most cases this will change when 
you change the magnetization direction.

My recommendation: Do init_so for the first magn. direction (accept 
eventual changes of symmetry/splitted atomic position/...).

Then do immediately with this struct file another init_so with the other 
magnetization direction and again accept the changes.

For security: do again init_so with the original magn.direction. There 
should be no change in symmetry anymore !

(If you want to do more than 2 directions, maybe you do EVERYTHING in P1 
(or P-1 if symmetry allows) (split all equivalent positions (maybe 
except inversion) and label the atoms (Fe1, Fe2, ..).

Once you have a common low symmetry/setup for both magn. directions:

runsp -orb -p -ec ...

x lapw1 -up  /-dn     (NO -orb !, otherwise you apply orb twice !!!!

steps 5-7 of your first strategy.
Then change the magn.direction and do again steps 5-7
--------------
In summary: You should have a common symmetry for all 3 calculations 
(without so and with the 2 magn.directions.)

You must not apply   -orb   twice (in lapw1 as in the runsp calc. and 
then again in lapwso )

  Regards


Am 23.09.2022 um 09:45 schrieb Marcus Ekholm via Wien:
> Dear all,
> 
> I would like to ask for a clarification on the various strategies to 
> calculate magnetic anisotropy energy (MAE) using the force theorem with 
> DFT+U. We get values for the MAE in the order of 10^-7 eV with Wien2k, 
> while with FLEUR and Quantum Espresso we get 10^-4 eV (for a slab), so 
> somewhere there is an error.
> 
> Based on previous posts to the mailing list, we try the following:
> 
> ***** STRATEGY 1
> 
> 1. init
> 2. runsp_lapw -orb -p -ec 0.000001
> 3. init_so (which will change EMAX and set an appropriate kmesh)
> 
> 4. set magnetization in case.inso
> 
> 5. x lapwso -up -p -orb
> 6. x lapw2 -up -so -p and x lapw2 -dn -so -p
> 7. Add the two :SUM values in case.scf2up/dn
> 
> Repeat #4-#7 to get energy differences
> 
> 
> ***** STRATEGY 2
> 
> 1. init
> 2. init_so, but set case.inso to skip SO for all atoms.
> 3. runsp_lapw -orb -p -ec 0.000001 -so
> 
> 4. set magnetization in case.inso
> 
> 5. x lapw1 -up -c and x lapw1 -dn -c
> 6. x lapwso -up -p -orb
> 7. x lapw2 -up -so -p and x lapw2 -dn -so -p
> 8. Add the two :SUM values in case.scf2up/dn
> 
> Change magnetization direction in case.inso and repeat #6-#8.
> 
> ***
> 
> Do these strategies make sense?
> 
> Best regards
> 
> 
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

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