It seems to me that the input positions are wrong. For example O25 and Ti27 positions are: O 0.50000 0.85806 0.30060 Ti 0.50001 0.67902 0.33161
Their distance is about 1 A, whereas Ti-O bond length is, as far as I remember, typically twice. It is a good habit to visualise a structure (using for example XCrysDen) before attempting any calculation, because in typical cases where two or more atoms are too close to each other or even overlap the sci cycles would never (or quite hardly) converge. Moreover, What I would do is try first the simplest run (e.g. calculation=‘sci’), removing all flags that could/should be set only if you thing you need them (mixing_mode, diagonalization, mixing_ndim). conv_thr seems too high. Do you have norm conserving pseudo potentials? In the case remove ecutrho. The k-point sampling along x and y directions is most probably insufficient for obtaining converged results, I would start with a 4x4x1 k-point grid at least. Giovanni > On 7 Sep 2017, at 18:58, Andrey Chibisov <[email protected]> wrote: > > Dear Colleagues. > I am studing the TiO2(101) surface. But during the structure relaxation the > total energy do not converge. > I attach the input file. > > -- > Best regards, > Andrey Chibisov. Ph.D. > Numerical method of mathematical physics Laboratory, > Computational Center, Russian Academy of Sciences. > Khabarovsk, Russia > Web page: https://www.researchgate.net/profile/A_Chibisov > http://ru.linkedin.com/pub/andrey-chibisov/55/253/986/en > > <TiO2.in>_______________________________________________ > Pw_forum mailing list > [email protected] > http://pwscf.org/mailman/listinfo/pw_forum -- Giovanni Cantele, PhD CNR-SPIN c/o Dipartimento di Fisica Universita' di Napoli "Federico II" Complesso Universitario M. S. Angelo - Ed. 6 Via Cintia, I-80126, Napoli, Italy e-mail: [email protected] Phone: +39 081 676910 Skype contact: giocan74 ResearcherID: http://www.researcherid.com/rid/A-1951-2009 Web page: http://people.fisica.unina.it/~cantele
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
