[Pw_forum] Problem with pseudopotential

[Stefano de Gironcoli]

Dear Izaak,

The local part of the potential might be compensated by the non-local one ..

Actually, in order to avoid ghost-states, I think that the local part  
is made as repulsive as possible at the atomic sites while the  
non-local part should add an attractive contribution. So maybe what  
you see is only one part of the picture because you are missing the  
non local terms.

Having said that I don't know if this is actually the case in your system.

If you plot the average charge density does it look reasonable or do  
you see it spilling out of the slab ?

   stefano


Quoting Nicola Marzari <nicola.marzari at materials.ox.ac.uk>:

>
>
> Dear Izaak,
>
>
> it seems that in Cu.pbe-d-rrkjus.UPF the average electrostatic potential
> inside the slab (i.e. the average over one layer of the planar average)
> would be higher than the vacuum level, so on electrostatic
> considerations a test charge would want to run away. So, I'm very
> baffled. Any one else has some suggestion? The two input files do look
> identical.
>
> The fermi energy that you print is the one given by the slab calculation
> (not really a good choice, unless you have a thick slab)
> or is given by calculating in bulk cu the position of the fermi energy
> with respect to the layer-average of the planar avarage, and then using
> that info in the slab calculation to position the fermi energy of the
> bulk, and obtain the work function?
>
> Last - a 14% is the last of the worries - LDA and GGA seem to be all
> over the place (have a look at the paper I had cited).
>
> 				nicola
>
>
>
> On 3/30/11 4:53 PM, Izaak Williamson wrote:
>> First of all, thank you so much for your replies.
>>
>> One more question: in the pbedrrkjus plot, there are maxima at the
>> location of each atomic layer whereas the pbevanak plot has minima. What
>> would cause this? Is it a problem?
>>
>>
>> Both the calculations were done using exactly the same parameters but
>> different pseudopotentials. I get the following values for workfunctions.
>> Cu.pbe-d-rrkjus.UPF ---> 3.956405463 eV
>> Cu.pbe-n-van_ak.UPF ---> 4.042282570 eV
>> Experimental value ---> 4.59 eV
>>
>> I know that DFT-GGA calculations can underestimate the workfunctions of
>> metal surfaces, but are these acceptable values (almost a 14% error)?
>>
>> Thank you again for all your help.
>>
>>
>> --
>> Izaak Williamson
>> Research Assistant
>> Physics Department
>> Boise State University
>
>
> --
> ----------------------------------------------------------------------
> Prof Nicola Marzari    Department of Materials    University of Oxford
> Chair of Materials Modelling  Director, Materials Modelling Laboratory
> nicola.marzari at materials.ox.ac.uk     http://mml.materials.ox.ac.uk/NM
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>



----------------------------------------------------------------
   SISSA Webmail https://webmail.sissa.it/
   Powered by Horde http://www.horde.org/