[Wien] volume optimisation of Fe2O3

[Torsten Andersen]

Dear Chandrika,

you should be aware that the "optimized" volume depends strongly on the 
PARAMETER called the XC-Potential, since there is no universal XC-Pot 
that works equally well for everything. DFT usually defines some 
accurate terms that can be determined well, and the rest is put into the 
XC-Pot (and with "exact" exchange, this is hence called the C-Pot). 
Therefore, an XC-Pot that is the best for one material might not be the 
best option for another material, and you get large deviations from 
experimental data. What is good for Fe is not necessarily good for its 
oxides...

If you repeat the optimization with a different XC-Pot you might get 
better results.

Best regards,
Torsten Andersen.

Chandrika wrote:
> Thanks Gerhard and Torsten for the suggestions.
> I reduced the convergence criteria and could obtain the optimised value
> at -12.5% volume. Somehow the structure files were getting the x-y-z
> positions all wrong in the various volume files, when I used the more
> stringent criteria.
> On the basis of experimental values I was expecting the volume optimisation
> to show a minimum around say 5%.
> I studied total energies at different k points and found 600 as the optimum
> (as suggested in Optimisation Notes).
> Thank you for your help,
> Chandrika
> 
> ----- Original Message ----- 
> From: "Gerhard H Fecher" <fecher at uni-mainz.de>
> To: "A Mailing list for WIEN2k users" <wien at zeus.theochem.tuwien.ac.at>
> Sent: Monday, September 11, 2006 7:45 PM
> Subject: Re: [Wien] volume optimisation of Fe2O3
> 
> 
> I remember that I needed already more than 20 cycles for convergence
> at the experimental lattice parameter with more relaxed settings than you
> do,
> maybe you should increase it , to be shure that the calculations are really
> converged.
> What is the meaning of   k-points= 600 (optimized) ?
> 
> Ciao
> Gerhard
> 
> 
>>> k-points= 600 (optimized)
>>>
>>>*Without Spin – polarization:*
>>>
>>> run_lapw –I –i 20 –ec 0.00001 –cc 0.0001
>>>
>>>
>>>
>>>*Non spin-polarized but with spin-orbit interaction*:
>>>
>>>run_lapw –so –cc 0.0001.
>>>
>>>Thank you for your suggestions,
>>>
>>>Chandrika
>>>
>>>
> 
> _______________________________________________
> 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
> 

-- 
Dr. Torsten Andersen        TA-web: http://deep.at/myspace/
AG Hübner, Department of Physics, Kaiserslautern University
http://cmt.physik.uni-kl.de    http://www.physik.uni-kl.de/