Temperature and energy

Message

Danny · #1 Post by **Danny** » Mon Oct 20, 2008 3:19 pm

Hi,

I was wondering if the following is possible in vasp:
A static calculation of a cell at a non-zero temperature.
I want to use this to check if a defect atom becomes unstable/less stable in favor of the clean system with regard to a certain temperature.
For this I would do calculations at different temperatures for both systems, and their difference in E0 or F(at certain T) should change sign at some point.

Can this be done? or do I have to do some sort of MD calculation for this?

thx,
Danny

#2 Post by **admin** » Tue Oct 21, 2008 11:36 am

please note that vasp only accounts for the electronic T, i.e. the excitation of certain phonon modes at elevated T is not taken into account for. Therefore I doubt whether your problem can be solved easily with vasp.

Danny · #3 Post by **Danny** » Wed Oct 22, 2008 12:47 am

(a difficult solution with vasp would also be usefull

)

So, just to be sure, the E given for a MD calculation(with IBRION=0, SMASS= -3 and POTIM=5.0) at the end of the first ionic step can not be considered to include the thermal temperature of the system?

thx
Danny

#4 Post by **admin** » Wed Oct 22, 2008 8:44 am

if you do MD, T includes the electronic T (via the semaring width of the electronic levels) and the contribution due to the kinetic energy of the ions ( T=2*Ekinetic/(the degrees of freedom) ).