[gpaw-users] hund's rule for Ti and V atom

[Marcin Dulak]

Hi Kurt,

On 11/22/2013 08:19 AM, Kurt Lejaeghere wrote:
> Dear Jun
>
> I have little experience with atomic calculations in gpaw, but I have 
> noticed the same in VASP. It is indeed an effect of PBE, which is not 
> able to predict the correct ground state electron configuration. In 
> VASP I find:
>
> Ti 3d^2 4s^2: -2.421208 eV
> Ti 3d^3 4s^1: -2.620505 eV
>
> V  3d^3 4s^2: -3.163138 eV
> V  3d^4 4s^1: -3.585726 eV
DFT problems with isolated transition metal atoms have been mentioned 
several times on the list,
last time: 
https://listserv.fysik.dtu.dk/pipermail/gpaw-users/2013-March/002013.html

 From the abstract of the first article:
"Atomic ground states are usually degenerate. It is demonstrated that 
the density functionals for the exchange-correlation energy that are 
commonly used are not invariant over the set of ground state densities. ....
It is recommended that atomic ground states that are actually used to 
calculate heats of atomization are made explicit ...."

This is the reason why cohesive energies are usually not published for 
many transition metals solids,
and for disagreements in results in (not very careful) publications 
dealing with atomization energies of TM compounds.

Best regards,

Marcin
>
> Best regards
>
> Kurt
>
> Citeren jun yan <junyan at stanford.edu>:
>
>> Hi,
>>
>>   I calculated with gpaw a single Ti and V atom with hund=True in the 
>> calculation.
>> for Ti I got final magmom to be 4 while Ti shall have 3d^2 4s^2 
>> electronic shell configuration;
>> and for V I got magmom to be 5 while V has 3d^3 4s^2
>> The initial guess for the magnetic moments are correct due to hund's 
>> rule, but during the calculations, the magmons changed and finally 
>> became 4 and 5 for Ti and V, respectively.
>>
>> Does anyone have experience with that ? Does it mean DFT(PBE) can't 
>> give the correct magnetic moment for these two atoms ? or did I do 
>> something wrong ? Thanks in advance for reply,
>>
>> all the best,
>> Jun
>> _______________________________________________
>> gpaw-users mailing list
>> gpaw-users at listserv.fysik.dtu.dk
>> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>
>
>
> _______________________________________________
> gpaw-users mailing list
> gpaw-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>
>


-- 
***********************************
  
Marcin Dulak
Technical University of Denmark
Department of Physics
Building 307, Room 229
DK-2800 Kongens Lyngby
Denmark
Tel.: (+45) 4525 3157
Fax.: (+45) 4593 2399
email: Marcin.Dulak at fysik.dtu.dk

***********************************