Dear Prof. Hutter, <div><br /></div><div>Thanks a lot for your help. Using the GAPW method improved the situation significantly (see table below).</div><div><br /></div><div>Dear all, </div><div><br /></div><div>I would like to share the updated results with you in case it may be useful for some of you. </div><div>In the table below, Calculation 6 is the same calculation as shown in the table sent on January 8, 2026. I include it to show the connection to the previous results.</div><div><br /></div><div>The difference between Calculation 8 and Calculation 7 shows the effect of including smoothing when using the GPW method. </div><div>There is a notable change in the vibrational frequencies, but I would not say that the situation improved. </div><div><br /></div><div>The change from Calculation 8 to Calculation 9 shows the effect of changing the method from GPW to GAPW while keeping the smoothing. </div><div>This improved the result significantly in that the vibrational frequencies of the modes corresponding to rotations became a lot closer to zero. </div><div><br /></div><div><div>The change from Calculation 9 to Calculation 10 shows the effect of removing the smoothing while using the GAPW method. </div><div>This had a small influence on the vibrational frequencies. </div><div><br /></div><div>Best regards,</div><div>Torstein </div></div><div><br /></div><div><img alt="Screenshot 2026-02-03 225522.png" width="944px" height="729px" src="cid:d751d6c5-77f1-4fb6-bfa1-351ef38283ef" /></div><div><br /></div><div><br /></div><div><br /></div><div><br /></div><div><br /><br /></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">fredag 9. januar 2026 kl. 09:44:52 UTC+1 skrev Jürg Hutter:<br/></div><blockquote class="gmail_quote" style="margin: 0 0 0 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Hi
<br>This behaviour is related to the way CP2K integrates XC functionals. The integration is
<br>done directly on the PW grid (in real space). This integration causes both, translational and
<br>rotational invariance to be violated for molecules. Usually, the amount of translation and
<br>rotation seen depends on the functional and the pseudopotential.
<br>The effect on molecular vibrations is mostly small (see frequencies before and after removing
<br>of translation/rotation).
<br>
<br>1) You can use FULLY_PERIODIC T to remove 'rotations'
<br>2) You can use GAPW and/or smoothing methods in XC. This improves results somewhat.
<br>
<br>regards
<br>JH
<br>
<br>________________________________________
<br>From: <a href data-email-masked rel="nofollow">cp...@googlegroups.com</a> <<a href data-email-masked rel="nofollow">cp...@googlegroups.com</a>> on behalf of Torstein Fjermestad <<a href data-email-masked rel="nofollow">tfjer...@gmail.com</a>>
<br>Sent: Thursday, January 8, 2026 8:44 PM
<br>To: cp2k
<br>Subject: [CP2K:22038] Rotational modes of methanol have high frequencies (> 550 cm-1), and they do not look like rotations
<br>
<br>CP1K version: 2023.1
<br>
<br>Dear all,
<br>
<br>I am doing a frequency calculation on a methanol molecule in a cubic unit cell of length 20.0 Å, using the M06 functional.
<br>I have noticed that the three lowest frequencies, corresponding to rotational modes, are very high (> 550 cm-1, see table below). I am aware that the frequencies of rotational modes will not be exactly zero in a numerical calculation, but I was at least expecting frequencies lower than 100 cm-1 for these modes.
<br>Furthermore, the modes do not look like rotations, but rather like angle bending and twisting of a torsion.
<br>
<br>I investigated the convergence with respect to the following parameters:
<br>EPS_SCF, EPS_DEFAULT, CUTOFF, REL_CUTOFF (see table below), but it had a negligible effect on the vibrational frequencies.
<br>Are there other crucial parameters that may affect the vibrational frequencies that I am not aware of?
<br>Please find the output files attached (The input is given at the start of the *.out files).
<br>Thanks in advance for your help.
<br>
<br>Best regards,
<br>Torstein
<br>
<br>
<br>
<br>[Screenshot 2026-01-08 204106.png]
<br>
<br>
<br>
<br>
<br>--
<br>You received this message because you are subscribed to the Google Groups "cp2k" group.
<br>To unsubscribe from this group and stop receiving emails from it, send an email to <a href data-email-masked rel="nofollow">cp2k+uns...@googlegroups.com</a><mailto:<a href data-email-masked rel="nofollow">cp2k+uns...@googlegroups.com</a>>.
<br>To view this discussion visit <a href="https://groups.google.com/d/msgid/cp2k/27f8babb-efcf-4abe-9b7e-1a77a956e05an%40googlegroups.com" target="_blank" rel="nofollow" data-saferedirecturl="https://www.google.com/url?hl=no&q=https://groups.google.com/d/msgid/cp2k/27f8babb-efcf-4abe-9b7e-1a77a956e05an%2540googlegroups.com&source=gmail&ust=1770239478951000&usg=AOvVaw1-lxDvj5fLUDGc4WsfduDU">https://groups.google.com/d/msgid/cp2k/27f8babb-efcf-4abe-9b7e-1a77a956e05an%40googlegroups.com</a><<a href="https://groups.google.com/d/msgid/cp2k/27f8babb-efcf-4abe-9b7e-1a77a956e05an%40googlegroups.com?utm_medium=email&utm_source=footer" target="_blank" rel="nofollow" data-saferedirecturl="https://www.google.com/url?hl=no&q=https://groups.google.com/d/msgid/cp2k/27f8babb-efcf-4abe-9b7e-1a77a956e05an%2540googlegroups.com?utm_medium%3Demail%26utm_source%3Dfooter&source=gmail&ust=1770239478951000&usg=AOvVaw0H-zyiKrRCfLwROK5v8GhB">https://groups.google.com/d/msgid/cp2k/27f8babb-efcf-4abe-9b7e-1a77a956e05an%40googlegroups.com?utm_medium=email&utm_source=footer</a>>.
<br></blockquote></div>
<p></p>
-- <br />
You received this message because you are subscribed to the Google Groups "cp2k" group.<br />
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:cp2k+unsubscribe@googlegroups.com">cp2k+unsubscribe@googlegroups.com</a>.<br />
To view this discussion visit <a href="https://groups.google.com/d/msgid/cp2k/3fee0fdc-3615-40e4-be21-925ff1cfb9e4n%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/3fee0fdc-3615-40e4-be21-925ff1cfb9e4n%40googlegroups.com</a>.<br />