Question about vibrational analysis

70 views
Skip to first unread message

Chn

unread,
Jun 5, 2020, 3:35:39 AM6/5/20
to cp2k
Dear experts in CP2K,

I am a beginner of CP2K package and I am really thankful if my naive question can be explained.
The model is PdCd (111) with small molecule adsorbed on it. When performing a GEO_OPT calculation, I added the keyword "UKS" for spin-polarized calculation in the section "DFT", due to the transition metals that are Pd and Cd atoms in the system according to the tutorials online. However it requires me about three times higher times for the calculation. For the vibrational analysis of the adsorbed molecule, a very tight threshold is needed for "EPS_SCF" to get accurate low frequencies, it will take me too much time. I do not know weather the keyword "UKS" is still necessary here.
Many thanks in advance!

Regards,
chn

Thomas Kühne

unread,
Jun 7, 2020, 6:45:53 AM6/7/20
to 'Dorothea Golze' via cp2k
Dear Can, 

if not all vibrational modes are necessary, e.g. for the substrate, the best 
way to save computational time is to use the MODE_SELECTIVE option. 

Best, 
Thomas

--
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/b1960a87-e117-4ac0-a3b0-94da7efea13ao%40googlegroups.com.



==============================
Thomas D. Kühne
Dynamics of Condensed Matter
Chair of Theoretical Chemistry
University of Paderborn
Warburger Str. 100
D-33098 Paderborn
Germany

Chn

unread,
Jun 7, 2020, 9:04:20 PM6/7/20
to cp2k
Dear Thomas,

many thanks for your reply!

regards,
chn



在 2020年6月7日星期日 UTC+8下午6:45:53，tkuehne写道：
Dear Can, 

if not all vibrational modes are necessary, e.g. for the substrate, the best 
way to save computational time is to use the MODE_SELECTIVE option. 

Best, 
Thomas
Am 05.06.2020 um 09:35 schrieb Chn <chen...@gmail.com>:

Dear experts in CP2K,

I am a beginner of CP2K package and I am really thankful if my naive question can be explained.
The model is PdCd (111) with small molecule adsorbed on it. When performing a GEO_OPT calculation, I added the keyword "UKS" for spin-polarized calculation in the section "DFT", due to the transition metals that are Pd and Cd atoms in the system according to the tutorials online. However it requires me about three times higher times for the calculation. For the vibrational analysis of the adsorbed molecule, a very tight threshold is needed for "EPS_SCF" to get accurate low frequencies, it will take me too much time. I do not know weather the keyword "UKS" is still necessary here.
Many thanks in advance!

Regards,
chn

--
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 cp...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages