<div dir="ltr"><div dir="ltr"><div><br></div>Try one of the following:</div><div dir="ltr">- use the development version</div><div dir="ltr">- use smearing='gauss'.</div><div>Explanation here: <a href="https://gitlab.com/QEF/q-e/-/issues/81">https://gitlab.com/QEF/q-e/-/issues/81</a>. I am aware of cases of oscillations between two configurations that were caused by this (now solved) problem.<br></div><div><br></div><div>Paolo<br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, May 12, 2021 at 12:42 PM Mauro Sgroi <<a href="mailto:maurofrancesco.sgroi@gmail.com">maurofrancesco.sgroi@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>Dear users,</div><div>I'm writing to ask for help with a geometry optimization of Li2TiS3.</div><div>For a particular structure the BFGS algorithm gives the message
CASE: energy _new > energy _old and the optimization starts oscillating between the same energy values.
</div><div><br></div><div>For example:<br></div><div> number of scf cycles = 184<br> number of bfgs steps = 4<br><br> energy old = -3664.4342647062 Ry<br> energy new = -3664.4341932047 Ry<br><br> CASE: energy _new > energy _old<br><br> new trust radius = 0.0000060248 bohr<br><br> trust_radius < trust_radius_min<br><br> resetting bfgs history<br><br> new conv_thr = 1.0E-11 Ry<br></div><div> <br></div><div> number of scf cycles = 185<br> number of bfgs steps = 4<br><br> energy old = -3664.4342647062 Ry<br> energy new = -3664.4341596224 Ry<br><br> CASE: energy _new > energy _old<br><br> new trust radius = 0.0002779227 bohr<br> new conv_thr = 1.0E-11 Ry</div><div><br></div><div> number of scf cycles = 186<br> number of bfgs steps = 4<br><br> energy old = -3664.4342647062 Ry<br> energy new = -3664.4341932047 Ry<br><br> CASE: energy _new > energy _old<br><br> new trust radius = 0.0000060248 bohr<br><br> trust_radius < trust_radius_min<br><br> resetting bfgs history<br><br> new conv_thr = 1.0E-11 Ry</div><div><br></div><div> </div><div>My input is below. I played a bit with the values of the trust_radius min and max and also tried the damped optimization with no results.</div><div>Have you any suggestions to solve this issue?</div><div>Thanks a lot and best regards,</div><div>Mauro Sgroi.</div><div>Centro Ricerche FIAT</div><div><br></div><div>&CONTROL<br> title = 'Li2TiS3'<br> pseudo_dir = '/workhpc/FCA/FCA_CRF_STRUT/sgroi/DATABASE/ESPRESSO'<br> prefix = 'POSCAR-1012.xyz'<br> outdir = '/workhpc/FCA/FCA_CRF_STRUT/sgroi/tmp/LTS'<br> calculation = 'relax'<br> disk_io = 'none'<br> nstep = 500<br> /<br> &SYSTEM<br> ibrav = 0<br> A = 10.736<br> nat = 54<br> ntyp = 3<br> ecutwfc = 52<br> ecutrho = 575<br> occupations = 'smearing'<br> degauss = 0.005<br> smearing = 'mv'<br> /<br> &ELECTRONS<br> electron_maxstep = 100<br> conv_thr = 1.0D-9<br> diago_thr_init = 1e-4<br> startingpot = 'atomic'<br> startingwfc = 'atomic'<br> mixing_mode = 'plain'<br> mixing_beta = 0.1<br> diagonalization = 'david'<br> diago_david_ndim = 4<br> scf_must_converge = .false.<br> /<br>&IONS<br> ion_dynamics = 'bfgs'<br> trust_radius_max = 1.5<br> trust_radius_min = 1.0d-4<br>/<br><br>ATOMIC_SPECIES<br> Li 6.941 Li.pbesol-sl-kjpaw_psl.1.0.0.UPF<br> Ti 47.867 Ti.pbesol-spn-kjpaw_psl.1.0.0.UPF<br> S 32.06 S.pbesol-nl-kjpaw_psl.1.0.0.UPF<br><br>K_POINTS {gamma}<br><br>CELL_PARAMETERS {alat}<br>0.000000 0.707107 0.707107 <br>0.707107 0.000000 0.707107 <br>0.707107 0.707107 0.000000<br><br>ATOMIC_POSITIONS {CRYSTAL}<br>S 0.1633831501 0.1505176116 0.1670420977<br>S 0.8343669301 0.8304670782 0.8418548293<br>S 0.1790412476 0.1764413502 0.4935621000<br>S 0.8247343817 0.8408510709 0.4917316687<br>S 0.1694651936 0.5002232433 0.1698005825<br>S 0.8212707269 0.5090769817 0.8266090665<br>S 0.4974749523 0.1673871640 0.1804228455<br>S 0.4927868323 0.8404319567 0.8296118314<br>S 0.1692023862 0.8072918196 0.8335429994<br>S 0.4890640593 0.1774332300 0.5033543118<br>S 0.1641898391 0.8317001666 0.4917151005<br>S 0.4912509798 0.4872129048 0.1840572305<br>S 0.1641576299 0.4968050469 0.8418554032<br>S 0.8474555693 0.1703269784 0.1606609978<br>S 0.8507828797 0.1558342390 0.8390628857<br>S 0.1743639033 0.4835156874 0.5118083355<br>S 0.8447227515 0.1643048203 0.4777090676<br>S 0.1590903063 0.8641836879 0.1479836531<br>S 0.4727465739 0.1769467379 0.8347894587<br>S 0.8594550184 0.8282327233 0.1509046931<br>S 0.1615534801 0.1693308123 0.8476401199<br>S 0.8542522557 0.4928254752 0.1478553155<br>S 0.4891060282 0.8469926505 0.1578063062<br>S 0.5072663898 0.5105322763 0.8198128542<br>S 0.8244012240 0.5104213752 0.4987828905<br>S 0.4982413076 0.8198464677 0.5160373591<br>S 0.4783486409 0.4993651057 0.5240038320<br>Ti -0.0265989094 0.0242519853 0.0081914789<br>Ti 0.3631233375 -0.0215014503 -0.0102111071<br>Ti 0.0266113267 0.3269523470 0.0099648008<br>Ti -0.0075871154 -0.0172900760 0.3646703424<br>Ti 0.3419155890 0.3288881722 0.6436175035<br>Ti 0.3571094741 0.6656725341 0.6686450804<br>Ti 0.6930348083 0.6408751632 0.6601932583<br>Ti -0.0207406070 0.6850583681 -0.0154635777<br>Ti 0.3285981104 0.3393287536 0.3118356500<br>Li 0.6587856458 0.6678472981 0.3328700152<br>Li 0.6556429250 0.3430241928 0.3388163527<br>Li 0.3345541443 0.6611504368 0.3292467652<br>Li 0.6715595113 0.3269333260 0.6678239753<br>Li 0.0060226298 0.6605967591 0.3365283850<br>Li 0.0126944052 0.3286927896 0.3395337969<br>Li -0.0038974319 0.3373998592 0.6566343563<br>Li 0.6576999667 0.0033507055 0.3346167570<br>Li 0.3246726313 0.0045111369 0.6656645737<br>Li 0.3330391120 0.0093228380 0.3367520122<br>Li 0.3419873713 0.3234714033 0.0165197646<br>Li 0.3395750477 0.6570272725 0.0135071994<br>Li 0.6551579591 0.3336210050 0.0149205898<br>Li 0.6653074562 0.0041400082 0.0044649389<br>Li 0.0073038566 0.0092546868 0.6496021276<br>Li 0.6635360145 0.0085920982 0.6662860686<br>Li 0.6489610883 0.6668026630 0.0048670464<br>Li -0.0102428937 0.6735271531 0.6598841003<br></div><div><br></div></div>
_______________________________________________<br>
Quantum ESPRESSO is supported by MaX (<a href="http://www.max-centre.eu" rel="noreferrer" target="_blank">www.max-centre.eu</a>)<br>
users mailing list <a href="mailto:users@lists.quantum-espresso.org" target="_blank">users@lists.quantum-espresso.org</a><br>
<a href="https://lists.quantum-espresso.org/mailman/listinfo/users" rel="noreferrer" target="_blank">https://lists.quantum-espresso.org/mailman/listinfo/users</a></blockquote></div><br clear="all"><br>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div>Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,<br>Univ. Udine, via delle Scienze 206, 33100 Udine, Italy<br>Phone +39-0432-558216, fax +39-0432-558222<br><br></div></div></div></div></div></div>