Stress Tensor

Message

pmignon · #1 Post by **pmignon** » Mon Jul 07, 2008 3:17 pm

Dear users,

how the stress tensor is calculated in VASP ? Does it contains both the kinetic and the potential part of the stress tensor ?
If not, the kinetic part of the stress tensor is
SUM (over the i atoms) Mi * Vi,a * Vi,b
Vi,a is the velocity component of atom i in the a direction and Mi is the atomic mass of atom i.
If I want to include this component in the stress tensor, Can I use the velocities from the CONTCAR file?

Thank you ,
Pierre.

#2 Post by **admin** » Mon Jul 07, 2008 6:03 pm

please have a look at OUTCAR (the lines above the one in which the total stress is written): the Local, n-local and augmentation contributions stem from the potential (the respective contributions to the PP)

pmignon · #3 Post by **pmignon** » Mon Jul 07, 2008 9:49 pm

Thank you for the fast reply.
Indeed I did not see that all contributions (including the kinetic) were summed to get the stress tensor. However looking to the code it seems to me that this kinetic term comes from "the change in the kinetic energy of the plane wave basis states as the size of the cell changes formula (10.50) in thesis gK"
This is not related to the momentum/velocity of the atoms.
I am not an expert and I may be wrong ... ?
By the way is there any possibilities to get an electronic version of the PhD thesis of Georg Kress ?
Thank you,
Pierre.