[ase-users] Dimer

[Michael Joseph Waters]

Hi Zainab,

You may want to try changing your optimizer. The documentation lists: BFGS, BFGSLineSearch, LBFGS, LBFGSLineSearch, GPMin, MDMin and FIRE as options. Many of the optimizers have options for maxstep/maxmove which limit large position jumps. Some optimizers have a damping option which can help.


Are you using molecular mechanics or an ab initio method like DFT for your dimer calculations?

If you are using molecular mechanics, try extending your cut-off distances.

If you are using an ab initio method, increasing your spatial sampling/plane wave cutoff and using tighter energy tolerances can help. If you are working with a spin-polarized ab initio method, things can get trickier due to the coexistence of multiple spin PES.

Best,
-Mike
________________________________
From: ase-users-bounces at listserv.fysik.dtu.dk <ase-users-bounces at listserv.fysik.dtu.dk> on behalf of Alaithan, Zainab via ase-users <ase-users at listserv.fysik.dtu.dk>
Sent: Thursday, January 23, 2020 7:06 AM
To: Ase-users <ase-users at listserv.fysik.dtu.dk>
Subject: [ase-users] Dimer

Dear Ase-user,

Can someone please share some tips about how to restart the dimer calculation from the dimer trajectory? and what is the appropriate displacement vector in this case? especially if the dimer calculation was converging and I just want to tighten the convergence criteria.

I will appreciate any help.

Regards,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20200123/8d729be0/attachment-0001.html>