Hi Marcella<div><br></div><div><br></div><div>OK. Thank you very much.</div><div><br></div><div><br></div><div><br></div><div>Yanyan</div><div><br></div><div><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">在2021年7月19日星期一 UTC+8 下午7:05:30<Marcella Iannuzzi> 写道：<br/></div><blockquote class="gmail_quote" style="margin: 0 0 0 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><br>Hi Yanyan,<div><br></div><div>One way to replicate the simulation box is via the MULTIPLE_UNIT_CELL keyword in both the CELL section and the TOPOLOGY section.</div><div>However, if you are simulating a defective structure that is not optimal, because you would replicate the defect as well, creating an array of defects. </div><div>You should first replicate the simulation box and coordinates of the pristine bulk a proper number of times and then create the defect. </div><div>How many times is a proper number of times depends on your system.</div><div><br></div><div>Regards</div><div>Marcella</div><div><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Saturday, July 17, 2021 at 8:36:28 AM UTC+2 <a href data-email-masked rel="nofollow">ya...@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
Marcella<div><br></div><div>Yes. Thanks for your advice.</div><div><br></div><div>And I found another question when I tried to calculate NV- center, Spin contamination estimate was not implemented for k-points. </div><div>So I should broaden the supercell. How much should I set the length? Now the length of the supercell is 7.5A*7.5A*7.5A</div><div><br></div><div> *** WARNING in qs_scf_post_gpw.F:1776 :: Spin contamination estimate not ***<br> *** implemented for k-points. ***<br></div><div><br>Yanyan</div><div><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">在2021年7月15日星期四 UTC+8 下午5:22:36<Marcella Iannuzzi> 写道：<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 ...<div><br></div><div>Most probably you did not activate LSD in your GEO_OPT run. </div><div>This is what one would deduce from the error message that you report.</div><div>Regards</div><div>Marcella<br><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Wednesday, July 14, 2021 at 4:45:30 AM UTC+2 <a rel="nofollow">ya...@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 All<div><br></div><div>I tried to optimize a O2 molecule (triplet) when MULTIPLICITY=1 or 3, respectively. And the result is that their energies were equal. </div><div>So I used their wfn file to calculate the energy finely. But an error occurred when I used the wfn file whose MULTIPLICITY=3. </div><div> *******************************************************************************<br> * ___ *<br> * / \ *<br> * [ABORT] *<br> * \___/ Restarting an LSD calculation from an LDA wfn only works for *<br> * | multiplicity=1 (singlets). *<br> * O/| *<br> * /| | *<br> * / \ qs_mo_io.F:934 *<br> *******************************************************************************<br></div><div><br></div><div>Then I closed the SCF_GUASS RESTART, the calculation completed successfully. </div><div>and the energy of MULTIPLICITY=3 was really lower than MULTIPLICITY=1.</div><div><br></div><div>So I want to confirm if the MULTIPLICITY setting doesn’t work in GEO_OPT?</div><div><br></div><div><br></div><div><br></div><div><br></div></blockquote></div></blockquote></div></blockquote></div></blockquote></div>