<div><br></div><div><br></div>Dear Alex<div><br></div><div>See section 2.4 in https://doi.org/10.1016/j.cpc.2004.12.014.</div><div>Improvements are obtained by applying smoothening (see <a href="https://manual.cp2k.org/cp2k-9_1-branch/CP2K_INPUT.html">CP2K_INPUT</a> / <a href="https://manual.cp2k.org/cp2k-9_1-branch/CP2K_INPUT/FORCE_EVAL.html">FORCE_EVAL</a> / <a href="https://manual.cp2k.org/cp2k-9_1-branch/CP2K_INPUT/FORCE_EVAL/DFT.html">DFT</a> / <a href="https://manual.cp2k.org/cp2k-9_1-branch/CP2K_INPUT/FORCE_EVAL/DFT/XC.html">XC</a> / <a href="https://manual.cp2k.org/cp2k-9_1-branch/CP2K_INPUT/FORCE_EVAL/DFT/XC/XC_GRID.html">XC_GRID</a> )</div><div>or using GAPW</div><div><br></div><div>Regards</div><div>Marcella</div><div><br></div><div><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Saturday, June 11, 2022 at 10:21:50 PM UTC+2 alexvd...@gmail.com wrote:<br/></div><blockquote class="gmail_quote" style="margin: 0 0 0 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><div>Hi Matt,</div><div>Thanks for the feedback. Which XC functionals are particularly susceptive to this effect? Do you have some references I could look into for this? I tried increasing the PW cutoff, but it is quite costly, and makes it difficult for me to do a high-throughput screening experiment.<br>Best,</div><div>Alex<br></div><br><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Friday, 10 June 2022 at 16:13:41 UTC-4 <a href data-email-masked rel="nofollow">mattwa...@gmail.com</a> wrote:<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 Alex,<div>this isn't to do with electrostatics as such, I don't think - it is something called the eggbox effect. CP2K projects the density onto a regular grid to calculate electrostatics and XC energy and this projection is not translationally invariant as you move atoms across the cell. The variations you see are probably due to the effect being cancelled out to some extent in some cases.</div><div>The simplest solution is to increase the PW cutoff in the MGRID section to alleviate the issue. It also depends on the XC functional used.</div><div>Matt<br><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Friday, 10 June 2022 at 20:16:23 UTC+1 <a rel="nofollow">alexvd...@gmail.com</a> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div>Sorry, I should clarify the files attached. In Z-align, the CO
molecule is aligned with the Z axis; Zperp is perpendicular to the Z
axis. In the *closepack files, the PBC is 1x1x10, whereas in the others
it is 5x5x10. The results are in the plot folder; where the figure
Z-align-closepack.txt.png plots the energy difference as a function of Z
displacement; a variation of 0.12 Hartrees is seen. In all the other
experiments, the variation is much smaller, 0.001 Hartrees. <br></div><div><br></div><div>Best,</div><div>Alex</div><br><br><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Friday, 10 June 2022 at 15:09:18 UTC-4 alex van der waals wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div>Hi,</div><div><br></div><div>I've been seeing a strange effect from a calculation. I have a single CO molecule, that I am displacing through the Z axis of a 5x5x10, 2x2x10 and a 1x1x10 PBC (with periodicity on for all axes, angles=90 degrees for all), and doing single point energy calculations for each. I expect that because its periodic, there should be no difference in energy where in the PBC I put the CO molecule. Strangely however, there is an energy difference; as I scan the Z axis, the energy reported fluctuates up and down periodically. I am using the PBE functional and a cutoff=300 fwiw.<br></div><div><br></div><div>I have attached figures and input files for the calcultions below.</div><div><br>Is this a known phenomenon?</div><div>Best,</div><div>Alex<br></div></blockquote></div></blockquote></div></blockquote></div></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 on the web visit <a href="https://groups.google.com/d/msgid/cp2k/a4659a35-0661-4317-b52b-267df1ff9dc4n%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/a4659a35-0661-4317-b52b-267df1ff9dc4n%40googlegroups.com</a>.<br />