You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to cp2k
Dear all,
I'm trying to calculate the frequencies of an organic molecule, and each time I get two or three negative values. I don't understand why. When I do the same thing with Gaussian, all the values are positive.
Can you help me? Thank you.
Frederick Stein
unread,
Mar 14, 2025, 5:34:10 AM3/14/25
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to cp2k
Dear Sihem,
Can you provide more details on your computational setup? Have you optimized all parameters of your setup (box size, CUTOFF, EPS_DEFAULT, ...)? Had you run a geometry optimization before you started the frequency calculation with CP2K?
Best,
Frederick
Jürg Hutter
unread,
Mar 14, 2025, 5:49:02 AM3/14/25
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to cp...@googlegroups.com
Hi
are you running using a supercell (periodic) or have you set
PERIODIC NONE for CELL and POISSON Solver ?
Default is to only remove 3 translations in periodic calculations,
no rotations will be removed (3 further frequencies).
Dear all,
I'm trying to calculate the frequencies of an organic molecule, and each time I get two or three negative values. I don't understand why. When I do the same thing with Gaussian, all the values are positive. Can you help me? Thank you.
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to cp2k
Dear Jürg and Frederick,
Thank you very much for your reply.
Yes, I perform a geometry optimization calculation before starting the frequency calculation, optimizing all the parameters.
In response to Jürg : I perform an optimization calculation on an isolated organic molecule, setting PERIODIC NONE and using POISSON_SOLVER ANALYTIC, followed by a frequency calculation.
For now, I'm only studying the isolated molecule; the periodic system will follow as soon as I understand how it works. The coordinates in the input file are obtained from a geometry optimization calculation performed with Gaussian.
I took the coordinates as-is from the Gaussian output file.
I have attached all the files.Thanks again for your help.
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to cp2k
Hi Sihem,
I checked your geometry and your forces. Is your system expected to be planar? If yes, try to set the y-coordinates of your atoms to 0.
Generally, a vibrational analysis requires tighter thresholds. Your forces need to have an higher accuracy than the force differences which are used by CP2K to run the analysis. With your value of the DX parameter, I see from your force differences that you need an accuracy significantly better than 1e-4 to 1e-5, i.e. at least 1e-6 to 1e-7. Especially an EPS_SCF of 1e-6 is usually to high. Please double-check the convergence of the parameters EPS_DEFAULT, EPS_SCF and CUTOFF and in your case maybe even the box-size in x-direction.
Best,
Frederick
Sihem AZIZI
unread,
Mar 17, 2025, 4:32:29 AM3/17/25
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to cp2k
Good evening Frederick,
Thanks again.
No, the molecule is not planar.
I obtained the molecule's coordinates by performing a geometry optimization with Gaussian.
Do I need to use another software to draw the molecule and get the xyz coordinates? For the rest, I'll try it out.
Thanks again,
Sihem
Jürg Hutter
unread,
Mar 17, 2025, 4:34:44 AM3/17/25
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to cp...@googlegroups.com
Hi
you have to reoptimize the structure in CP2K with the identical settings as will be used for
the vibrational analysis.