Implicit solvent calculation cannot get converged

li bingxin

unread,

Jul 27, 2020, 5:37:47 AM7/27/20

to cp2k

Dear CP2K users,

Recently I am doing a MgO slab calculation in implicit water using SCCS block of CP2K. I used the Andreussi Method and chose the CD5 as the derivative method because it requires lower CUTTOFF. However, it seems that the SCF loops cannot get converged and do not show any tendency to get converged. I have tried changing the mixing method from BROYDEN Mixing to PULAY Mixing and also using smaller mixing alpha value (0.2,0.1). However, these methods did not work in this case.

I have attached the input file as well as the output file here.

Could anyone please help me on that? I am looking forward to your reply.

Best Regards,

Bingxin

MgO_100.inp

MgO_100.out

Krack Matthias (PSI)

unread,

Jul 28, 2020, 5:37:00 AM7/28/20

to cp...@googlegroups.com

Hi Bingxin

SCCS is not implemented with k points in CP2K.

Matthias

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/14aef0b3-8bbf-4af2-af7d-e8bfab3df6dfo%40googlegroups.com.

li bingxin

unread,

Jul 28, 2020, 5:39:19 AM7/28/20

to cp...@googlegroups.com

Hi Matthias,

Thanks for your reply. Is OT method more suitable for the SCCS calculations?

Best,

Bingxin

Krack Matthias (PSI) <matthia...@psi.ch>于2020年7月28日周二10:37写道：

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/cda6b20bee114b75b16053be38cf2130%40psi.ch.

Krack Matthias (PSI)

unread,

Jul 28, 2020, 5:52:24 AM7/28/20

to cp...@googlegroups.com

No, the choice of an appropriate wavefunction optimizer depends rather on system type and setup, but OT is usually faster, especially for larger systems, and thus should be employed if possible.

M.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CAGYY63w6pQ-Z5SxUqwwRNuqHrprqaQqFo5FupV%2BwduYbzqdVUA%40mail.gmail.com.

li bingxin

unread,

Jul 28, 2020, 5:54:16 AM7/28/20

to cp...@googlegroups.com

Dear Matthias,

Thanks for your help!

Best,

Bingxin

Krack Matthias (PSI) <matthia...@psi.ch>于2020年7月28日周二10:52写道：

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/4c2b4b80a5184c019613369f30f33cf7%40psi.ch.

Reply all

Reply to author

Forward