The ABINIT Users Mailing List ( CLOSED )

[Xavier Gonze <gonze@pcpm.ucl.ac.be>]

Dear Marcel,

On 02 Feb 2007, at 16:08, Marcel Mohr wrote:

> Dear Abinit users
>
> I want to relax my structure regarding cell size. To do so I need  
> to set a non-zero ecutsm.
> When I enter the obtained relaxed coordinates in RF calculations
> I use ecutsm=0 (I think the manual says I have to use it for phonon  
> calculations).

In earlier versions of ABINIT you could not use ecutsm for response- 
function calculations.
For several versions now (two-three years ?) you can use ecutsm . See  
tests/v2#96,97,98 .
If the documentation still says that you cannot use ecutsm for  
phonon, then it is erroneous.
In this case, could you transmit in which file you saw this (thanks  
for your help !).

> Then stress occurs, about 4 magnitudes more on the level of 50 GPa.  
> I guess that introduces some error on the results.
>
> Any suggestions?
> Otherwise I have to do some a,c over energy curves to get the  
> relaxed coordinates.

So, try with ecutsm...

Regards,
Xavier