[Wien] Properties of corresponding atoms in afm calculations

Wed Sep 13 10:04:05 CEST 2006

Hi,
I checked the points you made.

Peter Blaha wrote:

>Hi,
>
>  
>
>>runafm seems not to be the problem. After an afm-cycle runsp does not 
>>change anything significantly.
>>    
>>
>
>What about the forces after runsp ? Do they become "equivalent" or stay
>different as obtained from runafm ?
>
>  
>
The forces stay different at similar values.

>What does "different forces" mean ? Could that be "numerical noise"
>(like forces of +20 and +21), or is it more like +20 and +10 ??
>
>  
>
You may call it "numerical noise" (it's like 10.276 vs. 10.400 etc.). 
But in my opinion this must not happen in an afm calculation since the 
down electrons are identical to the up electrons (by clmcopy). Thus 
everything that happens on atom A must be _exactly_ the same on atom A' 
(maybe with different sign, but same absolute value).
The different forces are quite inconvenient when doing a minimization as 
pointed out before.

>In the "noise" case, your procedure should be fine (eventually you may need
>better scf-convergence ?), otherwise something more
>serious may happen.
>
>  
>
I think the scf-convergence is okay (-ec 0.00001 -cc 0.00001). By the 
way this has no influence on the afm argument that properties of 
corresponding atoms should be the same.

>Please, also check carefully if the moments are really AF, or if you run
>into an ferrimagnetic solution.
>
>  
>
>A typical situation of the latter would be "bcc Mn", which is NOT
>antiferromagnetic, but the two Mn atoms get a slightly different moment.
>In such a situation runafm and runsp will naturally lead to differences. 
>
>
>  
>
runsp leads to the same total magnetic moment of 0.00000 as runafm. So I 
think the compound should be really antiferromagnetic.

Has anybody else observed this problem? If not I have to check again my 
inputs (although I did it now about five times).

Any further suggestions/comments are appreciated.
Thank you!

Ulrich Keßler

>  
>
>>My problem is that the forces of the corresponding atoms are not the 
>>same. During a minimization cycle this leads to different shifts of the 
>>corresponding atoms so that they are not any longer symmetry equivalent 
>>in the ferromagnetic supergroup. At the moment I stop the miminization 
>>after some cycles and edit the new struct file so that corresponding 
>>atoms possess equivalent sites. In my opinion the equivalent atoms in 
>>the supercell should have the same forces and should be shifted by the 
>>same amount.
>>
>>This is what I did:
>>I constructed the afm supercell in P1121/c (SG 14) containing 12 
>>independent atoms from the ferromagnetic structure in the supergroup 
>>B112/b (SG 15, 6 independent atoms). Since this is a klassegleiche 
>>subgroup the unit cell dimensions did not change. Every two atoms of  
>>the afm structure in P21/c correspond and possess positions which are 
>>equivalent in C2/c. During afminput I had to give the symmetry relation 
>>between these corresponding atoms (simply 0.5, 0, 0.5 since the B 
>>centering vanishes). After that runafm worked without problems except 
>>for the different forces of the corresponding atoms (I did not check 
>>other atom dependent properties).
>>
>>Is this something to live with (some kind of rounding error) or 
>>something more serious? And does an afm calculatiom make sense when the 
>>corresponding atoms are not mapped onto each other exactly by 
>>case.inclmcopy?
>>
>>If you need more information (struct-files etc.) please let me know.
>>
>>Thank you for your help,
>>Ulrich Keßler
>>
>>Peter Blaha wrote:
>>
>>    
>>
>>>runafm and runsp should yield the same results, in particular when 
>>>continuing with runsp AFTER a runafm convergence results must be 
>>>identical. (The could be cases when 
>>>runsp does not find and afm state, but eg. 2 moments which are different 
>>>(ferrimagnetic),..)
>>>
>>>Maybe you have a problem with your struct file (symmetry) or with afminput 
>>>(case.inclmcopy ?) ?
>>>
>>>
>>>
>>> 
>>>
>>>      
>>>
>>>>when calculating an antiferromagnetic compound using the runafm_lapw 
>>>>script, the symmetry relations between the antiferromagnetic atoms is 
>>>>exploited (they share the same site in the ferromagnetic supergroup). 
>>>>Thus I would expect that all properties of the corresponding atoms 
>>>>should be the same, especially the forces. However, I find that the 
>>>>forces of corresponding atoms tend to diverge in the course of an 
>>>>afm-SCF-cycle (not much, but significantly). Even more so when 
>>>>performing a minimization of positions.
>>>>
>>>>Is the reason some rounding error?
>>>>Is this serious? If yes, what could be done about it?
>>>>
>>>>Any comments are appreciated. Thank you.
>>>>
>>>>Ulrich Keßler
>>>>
>>>>
>>>>Dr. Ulrich Keßler
>>>>Institut für Anorganische Chemie
>>>>Universität Bonn
>>>>Gerhard-Domagk-Str. 1
>>>>D-53121 Bonn
>>>>Germany
>>>>Tel.: (+49) (0)228/73-5334
>>>>Fax: (+49) (0)228/73-5660
>>>>_______________________________________________
>>>>Wien mailing list
>>>>Wien at zeus.theochem.tuwien.ac.at
>>>>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>>>
>>>>   
>>>>
>>>>        
>>>>
>>>                                     P.Blaha
>>>--------------------------------------------------------------------------
>>>Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
>>>Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
>>>Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
>>>--------------------------------------------------------------------------
>>>_______________________________________________
>>>Wien mailing list
>>>Wien at zeus.theochem.tuwien.ac.at
>>>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>>
>>>
>>> 
>>>
>>>      
>>>
>>_______________________________________________
>>Wien mailing list
>>Wien at zeus.theochem.tuwien.ac.at
>>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>>    
>>
>
>
>                                      P.Blaha
>--------------------------------------------------------------------------
>Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
>Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
>Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
>--------------------------------------------------------------------------
>_______________________________________________
>Wien mailing list
>Wien at zeus.theochem.tuwien.ac.at
>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>
>  
>