[Pw_forum] Geometry Optimization

[Lorenzo Paulatto]

In data 06 agosto 2009 alle ore 10:54:12, MAHASIN ALAM  
<mahasinalam at yahoo.co.uk> ha scritto:
> But if I have a system with vacuum cell and if I need to relax the ions  
> as well as vacuum cell, can I do it using PWSCF? If yes can any one tell  
> me what I have to set in calculation type? Thank you for your help.

You can use vc-relax to relax both the ions and the unit cell (mainatining  
its symmetry); but it does NOT do what you want.

What you want to do is find the smaller amount of vacuum that guarantees a  
negligible interaction between periodic copies along the vacuum direction.  
This is equivalent to minimize forces between peridic images, but does not  
keep stress in account and will likely /increase/ the total energy.

On the other hand, vc-relax only tries to minimize stress and minimize  
energy. In practice, you have to do a few tryes with different amounts of  
vacuum and see when the total energy becomes stable. This may reuire as  
little as 5 Angstrom of empty space, for a simple slab geometry, but may  
be much more if you have dipolar molecules.

best regards

-- 
Lorenzo Paulatto
SISSA  &  DEMOCRITOS (Trieste)
phone: +39 040 3787 511
skype: paulatz
www:   http://people.sissa.it/~paulatto/

     *** save italian brains ***
  http://saveitalianbrains.wordpress.com/