Dear Elie Moujaes
Sorry for posting again. >(1) is changing the mesh in the DOS program means decreasing DeltaE? I'm sorry I wrote a little confusing description. I meant to say that, in my previous posting, the checking of the convergence in Fermi-energy with respect to the mesh of k-points in nscf calculation". You will/should observe the convergence as the increase of k-mesh. After you get the sufficient condition in the k-mesh, you will obtain more reliable value of Fermi-energy in DOS calculation. (note that, in my opinion, if you want to get the precise value as much as possible, the decision of the value of Fermi-energy should be done by counting number of electrons with looking output file obtained by DOS calculation (by the checking of integrated DOS). >(2) Concerning the convergence against smearing, does this mean that one has to decrease degauss (mine is degauss=0.01) to get better results? If you already checked the convergence of, for example, the total energy in your system against change of degauss ( and the number of k-points), I think it is no problem. ( Note: Your "degauss =0.01" is about 0.13 eV which value is within the error of the value of Fermi-energy which you reported. So I'm a little worried about the value of degauss. Did you check the change of Fermi-energy against the "degauss" ? Perhaps, my comments to you are incorrect ( if so, it lead the waste of time in your life, I want to say very sorry for confusing you, in advance). I hope you can get better result. Haruhiko Dekura ======================================== Haruhiko Dekura Postdoctoral Fellow Senior Research Fellow Center, Ehime University, 2?5 Bunkyo-cho, Matsuyama, Ehime 790-8577, Japan ===================================== > > > Thank you very much for your help. I have just got some few more questions: > > (1) is changing the mesh in the DOS program means decreasing DeltaE? > > (2) Concerning the convergence against smearing, does this mean that one > has to decrease degauss (mine is degauss=0.01) to get better results? > > You will find below a part of the bands input file. > > &control > prefix='GBdensity', > calculation='bands', > restart_mode='from_scratch', > pseudo_dir = > '/exp/home/caiapo/emoujaes/espresso/espresso-4.1.3/pseudo/', > outdir='/exp/home/caiapo/emoujaes/espresso/Moujaes-Results/', > > / > &system > ibrav= 0, celldm(1) =1.889725989, nat=60, ntyp= 1, ecutwfc =40.0, > ecutrho = 160.D0, occupations='smearing', smearing='mp', degauss=0.01 > / > &electrons > conv_thr=1.D-5, > mixing_beta=0.1D0, > > / > > CELL_PARAMETERS (alat) > 24.064488464 0.000772242 0.000000000 > -0.000547177 6.503051170 0.000000000 > 0.000000000 0.000000000 8.470514812 > > > ATOMIC_SPECIES > C 12.0107 C.pz-rrkjus.UPF > > ATOMIC_POSITIONS (angstrom) > C -11.330758616 -3.527803203 0.000000000 > C -10.659793092 -1.160339161 0.000000000 > C -12.039843315 -7.112619698 0.000000000 > C -12.041158182 -5.732701936 0.000000000 > C -10.941089654 -4.881442842 0.000000000 > C -10.331417950 -2.543878336 0.000000000 > C -8.972994221 -3.016349126 0.000000000 > C -8.181174559 -0.722870888 0.000000000 > C -9.522448653 -6.774832629 0.000000000 > C -9.676361200 -5.364937582 0.000000000 > C -8.645931928 -4.406311082 0.000000000 > > .......... > ........... > > > Thanks again > > Elie Moujaes > University of Nottingham > NG7 2RD > UK > > > > > > Date: Wed, 26 Jan 2011 01:42:39 +0900 > From: adadadad3018 at gmail.com > To: pw_forum at pwscf.org > CC: dekura at sci.ehime-u.ac.jp > Subject: Re: [Pw_forum] 60 atom Grain boundary... (Elie Moujaes) > > > > > > Dear Elie Moujaes > > > > > Do you mean that your calculated Fermi-energy in C60 system is not located > at which the DOS becomes zero even though you got the reasonable band > structure. > > > If so, I recommend a check for the convergence of Fermi-energy. > > For example, > > > please check the energy convergence against > > > (1) increase of the number of k-points in nscf calculation, after you > confirm that the KS energy spectrum is converged sufficiently in SCF > calculation (and also should check the convergence against the smearing > width) > > > (2) change in the mesh in DOS calculation. > > > > > May be you will get more reliable and better result. > > > > > Hope this helps. > > > > > Best Regards. > Haruhiko Dekura > > > > > > > ======================================== > > > Haruhiko Dekura > > > Postdoctoral Fellow > > > Senior Research Fellow Center, Ehime University, 2?5 Bunkyo-cho, Matsuyama, > Ehime 790-8577, Japan > ===================================== > > > > > > Dear all, > > I am trying to get the band structure and the DOS of a Grain boundary > system of 60 atoms. > > (1) The band structure seems ok but when I am trying to set the Fermi > energy to be the reference (i.e at 0) where two bands cross, it is not > working i.e. I am still getting that the two bands still cross at a negative > value (my Fermi energy is -0.098eV). Why is this happening? > > (2) The DOS of the system is not zero at zero energy (maybe this is a > consequence of 1) and it is a bit wiggly as well (see graph attached) > > > Thanks in advance > > Regards > > Elie Moujaes > University of Nottingham > NG7 2RD > UK > > _______________________________________________ Pw_forum mailing list > Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > http://www.democritos.it/pipermail/pw_forum/attachments/20110125/fae4cadc/attachment.htm > > ------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > End of Pw_forum Digest, Vol 43, Issue 49 > **************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20110126/aaa4e8f6/attachment.htm
