[Wien] Isolated Atomic Calculation - all time error

[Laurence Marks]

I strongly suggest that you use TEMPS with some reasonable broadening
such as 0.002 or perhaps 0.005 to get the calculation fairly well
converged first, and also make sure you understand how to do general
calculations first. For a single atom you can then reduce the
broadening for the final calculation. There are some cases for single
atoms where the problem is very ill-conditioned because it is
discontinuous at the solution. A better mixer will be available in the
next release, but even this can have problems, and the best one can do
is get an approximate solution good to perhaps 0.01eV (which is much
better than the accuracy of the functionals).

On 7/16/07, Stefaan Cottenier <Stefaan.Cottenier at fys.kuleuven.be> wrote:
>
> > HOW IS IT POSSIBLE TRUST TO BULK CALCULATIONS IF THERE IS NOT CLEAR
> > SCHEME FOR CALCULATION OF ONE ATOM UNTILL NOW?
>
> Don't start shouting...
>
> 1) There is a clear scheme (you are using it), but not all the defaults
> that are meant for normal solids work well for a highly artificial case
> as an almost empty supercell with a single free atom. You need to
> understand a bit more of the background of the method than is usually
> needed for a solid. No step-by-step scheme can subsitute for this
> understanding.
>
> 2) For a solid state code that works with crystals, a free atom *is*
> harder to calculate than a bulk solid...
>
> Stefaan
>
>
> Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>


-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
Web: www.numis.northwestern.edu
EMM2007 http://ns.crys.ras.ru/EMMM07/
Commission on Electron Diffraction of IUCR
www.numis.northwestern.edu/IUCR_CED