About SMEAR & MIXING

[Matt W]

You can search for "Fermi-Dirac smearing" in this forum, where there are a 
couple of suggestions on references. You might be best looking into 
suggestions from various plane wave codes that are more regularly used for 
simulating metals. From memory suggestions are from 0.1-0.3 eV across and 
down the transition metals / but you can easily find information on the 
'web.

The temperature you use depends on the density of states at/around (around 
being defined by the temperature you use) the Fermi level. Ideally you want 
to use as high a temperature (helps SCF convergence and, maybe, increases 
the accuracy of the gamma point only approximation) as possible, without 
really changing the properties of the material you are looking at...

HTH

Matt

On Monday, 4 February 2013 05:38:34 UTC, MR.Chem wrote:
>
> HEY, I AM TRYING TO CALCULATING COBALT SURFACE, AND USING SMEAR & MIXING 
> METHOD.  BUT I AM NOT SURE HOW TO SET UP "ELECTRONIC_TEMPERATURE [K]" .
> WHERE CAN I FIND THE TEMPERATURE? IS IT AROUND 2000K~3000K? DOES IT 
> CAN AFFECT THE RESULTS A LOT?
> ABOUT MIXING PART, I WROTE LIKE THIS:
> &MIXING 
>         METHOD BROYDEN_MIXING 
>         ALPHA   0.1 
>         BETA    1.5 
>         NBROYDEN  8 
>        &END
> IS THAT OK???
>  ALSO, CAN SOMEONE SUGGEST KINDA REFERENCES THAT INTRODUCING THIS METHOD, 
> PLEASE? 
>
> THANK YOU VERY MUCH FOR HELP!!!!
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20130206/962fdd03/attachment.htm>