dimer restart

Vasp transition state theory tools

Moderator: moderators

Post Reply
ngoldman
Posts: 4
Joined: Tue Aug 03, 2010 10:49 pm
Contact:

dimer restart

Post by ngoldman »

Hi Graeme et al.,

I'm having a strange issue with restarting my dimer calculation. The system is Au3 on a rutile TiO2 (110) surface. I'm using IOPT=2 with its default parameters, the default parameters for the dimer calculation, as well as EDIFF = 1E-7, EDIFFG = -1E-3, LREAL = F, ISYM = 0. Also, I'm restarting with CENTCAR and NEWMODECAR. At the end of one run, I have a force value of 0.04119, a torque value of 0.32725 and an angle of 0.05596. Upon restarting, I get 0.054166 for the force and 23.03281 (!) for the torque (I seem to have quit the calculation before the angle was calculated). Any idea of what's going on? My gut feeling was to switch to IOPT=1 since the forces are rather low.

Thanks,

Nir
ngoldman
Posts: 4
Joined: Tue Aug 03, 2010 10:49 pm
Contact:

Re: dimer restart

Post by ngoldman »

Let me correct that: I plan to switch to IOPT=3 (quick min).
graeme
Site Admin
Posts: 2290
Joined: Tue Apr 26, 2005 4:25 am
Contact:

Re: dimer restart

Post by graeme »

That doesn't sound right. If you tar.gz the runs, and attach them to the forum, I'll take a look and see if I can see something going wrong.

Quick check: did you copy the CENTCAR -> POSCAR and NEWMODECAR -> MODECAR?
ngoldman
Posts: 4
Joined: Tue Aug 03, 2010 10:49 pm
Contact:

Re: dimer restart

Post by ngoldman »

That's right -- I copied the CENTCAR -> POSCAR and NEWMODECAR -> MODECAR.

Unfortunately I can't readily give you a tar.gz copy of the run due to certain restrictions on the cluster I used. I'll attempt to replicate the error on another machine and send the files as soon as possible.

Thanks,

Nir
Post Reply