Page 1 of 1

Hessian matrix calculation

Posted: Mon Feb 25, 2008 8:09 pm
by Lecuyern
Hi everyone,

I am sorry to bring again the question of the Hessian matrix but the other posts did not totally answer my questions.

I need to make an Hessian matrix calculation for a system that include a membrane and an adsorbant.

I am totally new to setting manually the commands of VASP; I used to do it with the help of a graphic interface that unfortunately doesn't allow me to do this type of calculations.

So I would like to know exactly what are the VASP commands I need to include in the different input files in order to perform a correct Hessian Matrix calculation.

I saw in the other posts some commands like IBRION = 5 and NSW =1 (can it be more than one like 100?!) but what else is needed?

Could someone help me with this?

Thanks a lot

Hessian matrix calculation

Posted: Wed Feb 27, 2008 11:20 am
by admin
if you have in mind to calculate vibrational spectra, please rather
1) converge the ionic configuration of your system completely
2) calculate the spectrum for the converged geometry, using NSW=1.
As these runs are very time consuming (please note that the number of degrees of freedom - unless specified via the 'selective dynamics' tag in POSCAR - is 3*N(atoms)), because each ionic displacement has to be converged electronically
The number of considered displacements is not related to the NSW parameter for IBRION=5, but set to the number of degrees of freedom.
3) please reduce POTIM (to about 0.015) to make sure that each displacement is within the harmonic limit.