<html><body><div style="font-size: 12pt; font-family: times new roman, new york, times, serif; direction: null; color: #000000;" data-attr="forced_root_block_attrs">
<div style="font-size: 12pt; font-family: times new roman, new york, times, serif; direction: null; color: #000000;">
<div>Dear Prof. Giannozzi,</div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;"><br></div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;">Thank you for your response. I apologize for my late reply.</div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;"><br></div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;">I succeeded with the relaxation of my crystal structure achieving the following forces ans stress tensor.</div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs"> </div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs">
<div data-attr="forced_root_block_attrs"> Forces acting on atoms (cartesian axes, Ry/au):</div>
<div data-attr="forced_root_block_attrs"> atom 1 type 1 force = 0.00000000 0.00000000 0.00001488</div>
<div> atom 2 type 1 force = 0.00000000 0.00000000 -0.00001488</div>
<div> atom 3 type 2 force = 0.00000000 0.00000000 0.00000844</div>
<div> atom 4 type 2 force = 0.00000000 0.00000000 -0.00000844</div>
<div> atom 5 type 2 force = 0.00000000 0.00000000 -0.00001271</div>
<div> atom 6 type 2 force = 0.00000000 0.00000000 0.00001271</div>
<div> atom 7 type 2 force = 0.00000000 0.00000000 -0.00001894</div>
<div> atom 8 type 2 force = 0.00000000 0.00000000 0.00001894</div>
<div data-attr="forced_root_block_attrs"> Total force = 0.000040 Total SCF correction = 0.000000</div>
<div data-attr="forced_root_block_attrs"><br></div>
<div> Computing stress (Cartesian axis) and pressure</div>
<div data-attr="forced_root_block_attrs"> total stress (Ry/bohr**3) (kbar) P= -0.07</div>
<div> -0.00000070 0.00000000 -0.00000000 -0.10 0.00 -0.00</div>
<div> 0.00000000 -0.00000036 -0.00000000 0.00 -0.05 -0.00</div>
<div> -0.00000000 0.00000000 -0.00000034 -0.00 0.00 -0.05</div>
<div><br></div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs">However, when performing the phonons calculations at the Gamma point I obtained all negative frequencies:</div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs"> </div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs">
<div> freq ( 1) = -93.901745 [THz] = -3132.225047 [cm-1]</div>
<div> freq ( 2) = -88.011019 [THz] = -2935.731590 [cm-1]</div>
<div> freq ( 3) = -76.385627 [THz] = -2547.950243 [cm-1]</div>
<div> freq ( 4) = -76.150629 [THz] = -2540.111574 [cm-1]</div>
<div> freq ( 5) = -70.602435 [THz] = -2355.043727 [cm-1]</div>
<div> freq ( 6) = -70.549858 [THz] = -2353.289943 [cm-1]</div>
<div> freq ( 7) = -67.081504 [THz] = -2237.598108 [cm-1]</div>
<div> freq ( 8) = -66.962776 [THz] = -2233.637768 [cm-1]</div>
<div> freq ( 9) = -65.079559 [THz] = -2170.820413 [cm-1]</div>
<div> freq ( 10) = -64.752061 [THz] = -2159.896266 [cm-1]</div>
<div> freq ( 11) = -47.521629 [THz] = -1585.150933 [cm-1]</div>
<div> freq ( 12) = -47.342909 [THz] = -1579.189469 [cm-1]</div>
<div> freq ( 13) = -46.459335 [THz] = -1549.716620 [cm-1]</div>
<div> freq ( 14) = -46.436561 [THz] = -1548.956937 [cm-1]</div>
<div> freq ( 15) = -44.002439 [THz] = -1467.763369 [cm-1]</div>
<div> freq ( 16) = -43.980006 [THz] = -1467.015088 [cm-1]</div>
<div> freq ( 17) = -42.040378 [THz] = -1402.316054 [cm-1]</div>
<div> freq ( 18) = -42.036814 [THz] = -1402.197176 [cm-1]</div>
<div> freq ( 19) = -27.016208 [THz] = -901.163687 [cm-1]</div>
<div> freq ( 20) = -27.002908 [THz] = -900.720056 [cm-1]</div>
<div> freq ( 21) = -15.609053 [THz] = -520.661973 [cm-1]</div>
<div> freq ( 22) = -15.040852 [THz] = -501.708804 [cm-1]</div>
<div> freq ( 23) = -14.265259 [THz] = -475.837822 [cm-1]</div>
<div> freq ( 24) = -14.243345 [THz] = -475.106863 [cm-1]</div>
<div> </div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs">I checked the F.A.Q. page from Quantum Espresso but the frequencies are not small, thus I guess imposing ASR is not sufficient to figure out this problem.</div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs">I attach my scf file including the relaxed structure, I kindly ask you (or anyone in the community), to give some advice on what to do.</div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs"> </div>
<div style="font-size: 12pt; font-family: 'times new roman', 'new york', times, serif; color: #000000;" data-attr="forced_root_block_attrs">Thank you very much in advance. Best regards,</div>
</div>
</div>
<div><br></div>
<div id="signature-content-15e27e63-fb61-4390-9456-835d1575430a" data-marker="__SIG_PRE__">
<div>
<div><span style="font-size: 18pt;"><strong>Alejandro Lasso Castillo</strong></span><br><em>Erasmus Mundus Joint Master Student</em><br><em>Quantum Science and Technologies (QUARMEN)</em></div>
</div>
</div>
<div><br></div>
<div>
<div id="OLK_SRC_BODY_SECTION">
<div id="OLK_SRC_BODY_SECTION">
<blockquote style="margin: 0 0 0 .8em; border-left: 1px #ccc solid; padding-left: 1em;"><hr id="MESSAGE_DATA_MARKER"><strong>De: </strong>Paolo <paolo.giannozzi@uniud.it><br><strong>Para: </strong>Quantum <users@lists.quantum-espresso.org><br><strong>CC: </strong>Alejandro <alejandro.lasso-castillo@universite-paris-saclay.fr><br><strong>Fecha: </strong>miércoles, 12 de junio de 2024 18:28 CEST<br><strong>Asunto: </strong>Re: [QE-users] Relaxation on LaTe3<br><br>There might be a small mismatch, due to numerical reasons, between <br>forces and energies, in which case the BFGS minimization will never <br>reach the zero-force configuration. You may try to use damped dynamics <br>instead. BFGS uses both forces and energies, damped dynamics only forces. <br><br>In order to have very precise force values, one needs a highly converged <br>scf, but the scf threshold you have used (conv_thr) is very small so I <br>don't think this is the origin of the problem <br><br>Paolo <br><br>On 12/06/2024 16:59, Alejandro Lasso Castillo via users wrote: <br>> Dear users, <br>> <br>> I am currently trying to perform phonon calculations on LaTe3, and one <br>> keypoint in order to have good results is a very well relaxed structure. <br>> For this purpose I use a force threshold of 1d-05 (see input file attached). <br>> <br>> However, I find that it is very difficult to achieve this threshold (and <br>> it becomes harder when increasing the density of k-points). For a k-grid <br>> of 12 x 2 x 12 I try to relax the structure and eventually the bfgs <br>> algorithm gets stuck <br>> into the same three configurations. My strategy so far has been to <br>> reintroduce the best structure from the output into the input file and <br>> relaunch it but I can not achieve forces less than 6d-05. <br>> <br>> I ask you for some advice in order to achieve my threshold in case you <br>> have some experience with this kind of materials. <br>> <br>> *Alejandro Lasso Castillo* <br>> /Erasmus Mundus Joint Master Student/ <br>> /Quantum Science and Technologies (QUARMEN)/ <br>> <br>> <br>> _______________________________________________ <br>> The Quantum ESPRESSO community stands by the Ukrainian <br>> people and expresses its concerns about the devastating <br>> effects that the Russian military offensive has on their <br>> country and on the free and peaceful scientific, cultural, <br>> and economic cooperation amongst peoples <br>> _______________________________________________ <br>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu) <br>> users mailing list users@lists.quantum-espresso.org <br>> https://lists.quantum-espresso.org/mailman/listinfo/users <br><br>-- <br>Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche, <br>Univ. Udine, via delle Scienze 206, 33100 Udine Italy, +39-0432-558216 <br>X SCUOLA ENERGIE RINNOVABILI 16-21/9/2024 - PRE-ISCRIZIONE <br>https://agenda.infn.it/event/39573/ <br></blockquote>
</div>
</div></div>
</div>
</div></body></html>