Hi Giovanni, Thank you for your suggestions. I will increase the k-points and redo the calculations.
Francesco, you are absolutely right. Aftet putting no_overlap=.true that problem has gone.! Thank you to both of you! Manu University of Waterloo On Mon, Apr 18, 2016 at 4:15 AM, Francesco Pelizza < [email protected]> wrote: > Sorry to jump in this argument. I may be not of much help. > > For the question of overlapping bands or strange mixing of bands...In my > experience, bands become easier to be plotted if when you run bands.x you > add in the input file the flag: no_overlap= .true. > > That flag avoid eigenvalues overlapping and even plotband.x does a nice > job. > > Francesco > University of Strathclyde > > > > > On 18/04/16 08:56, Giovanni Cantele wrote: > > bands crossing are spurious and they are due to the fact that, after > running bands.x (or in the case you have not used it at all), the bands > order might be incorrect. That means that the n-th band (that is, the band > that at each k-point is > plotted as n-th eigenvalue) might actually contain, at different k-points, > energies belonging to different bands. In this case, if you not only plot > the points but also connect them with lines, you might see spurious > connection lines between > different bands. The way to cure this problem is to run bands.x before > plotband.x, but with a sufficient number of k-points (if they are too > distant bands.x might fail in following a band, that is, in assigning a > given eigenvalue to a band). > > Giovanni > > > > On 18 Apr 2016, at 05:34, Manu Hegde < <[email protected]> > [email protected]> wrote: > > Hi Giovanni, > > Thanks for your advice. > > I have calculated band structure for the zinc blende GaP lattice. When I > plotted using gnuplot it looks okay (plotgnu_gap.ps). But .ps format > from the plotband.x (plotband_gap.ps) does not look properly and there > some bands crossing. I am not sure why it is happening like this. I have > attached both the files. > > Regards, > Manu > > Dr.Manu Hegde > Postdoctoral Fellow > Department of Chemistry > University of Waterloo > > Waterloo, ON, N2L 3G1 > > > On Wed, Apr 6, 2016 at 6:26 AM, Giovanni Cantele < > <[email protected]>[email protected]> wrote: > >> Quite likely, you did not calculate the band structure along a path in >> the Brillouin zone, but used the output of the scf calculation (k-point >> grid). After running pw.x with calculation=‘scf’ or calculation=‘relax’, >> you need to rerun it with a new input file, calculation=‘bands’ and >> K_POINTS card containing a suitable path of k-points. See also examples of >> band structure calculation included in the QE-package. >> >> You can set Emin (Emax) to the value of the lowest (highest) eigenvalue), >> unless you need to select some specific energy range. >> >> Giovanni >> >> PS Users of this forum are usually kindly requested to sign messages with >> affiliation >> >> On 06 Apr 2016, at 05:25, Manu Hegde < <[email protected]> >> [email protected]> wrote: >> >> Hello All, >> >> I was trying to plot band structure of PbTiO3, I could only see dots in >> bands.ps file. I am not sure what is the reason for this. Is it due to >> the wrong k-point selection?. I am also not sure about energy selection >> along y-axis (Emax, Emin). I have attached bands.ps file please have a >> look. Please help to overcome from this problem. >> >> Thanks and Regards >> >> Manu >> <pbtio.bands.ps>_______________________________________________ >> 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.na.infn.it/~cantele >> >> >> _______________________________________________ >> Pw_forum mailing list >> [email protected] >> http://pwscf.org/mailman/listinfo/pw_forum >> > > <plotband_gap.ps><plotgnu_gap.ps> > _______________________________________________ > 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.na.infn.it/~cantele > > > > _______________________________________________ > Pw_forum mailing > [email protected]http://pwscf.org/mailman/listinfo/pw_forum > > > > _______________________________________________ > Pw_forum mailing list > [email protected] > http://pwscf.org/mailman/listinfo/pw_forum >
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
