[DFTB-Plus-User] band structure SCC convergence

R.Perez Garcia r.perez.garcia at student.rug.nl
Wed Jul 11 11:31:56 CEST 2012


Dear Mr. Knaup,
Thank you for the answer.
I see now.. I was trying to see a nice "typical" plot but I believe for my system it is just very messy plot.
I really appreciate your suggestions.
Best regards: R

On 11-07-12, "Jan M. Knaup"  <Jan.Knaup at bccms.uni-bremen.de> wrote:
> Dear Rodrigo,
> 
> 
> the band_tot.dat file is structured to be very convenient to plot. The first column is the k-point number and will serve as your x-axis. You can use the index to apply the usual \gamma, K, L etc labels corresponding to your input file. The other columns contain the energies for the eigenstates at the corresponding k-point.
> 
> 
> 
> 
> Practically, you can plot the bant_tot.dat file e.g. by importing it into xmgrace as an NXY data set. (data->import ascii-> change XY to NXY). Or you can use a command like
> "plot for [i =2:XXX]  'bant_tot.dat' u 1:i with lines notitle"
> 
> 
> in gnuplot 4.4 or higherm where you replace the XXX by the number of columns in the file (should be the same as (N_el/2)+1 ).
> 
> 
> Hope that helps,
> 
> 
> Jan
> 
> 
> 
> Jan M. Knaup                      | Fon +49-(0)421-218-62351
> Dipl. Phys. Dr. rer. nat.         | Fax +49-(0)421-218-62770
> 
> 
> Universität Bremen - BCCMS        | 
> Am Fallturm 1                     | Jan.Knaup at bccms.uni-bremen.de
> 
> 
> 28359 Bremen                      | JanKnaup at gmail.com
> 
> 
> Germany                           | www.bccms.uni-bremen.de(http://www.bccms.uni-bremen.de/)
> 
> 
> 
> 2012/7/11 R.Perez Garcia <r.perez.garcia at student.rug.nl>
> 
> > 
> > 
> > Dear dr. Penazzi,
> > Thank you very much for your kind answer.
> > I believe i follow the tutorial adecquately. I should copy the answer prof. Aradi gently send me a few seconds ago.
> > " If you use MaxSCCIterations = 1, the SCC won't converge, but you also
> > 
> > 
> > do not want it to converge, so it is fine.   Best regards,  Bálint"
> > So that problem is solved. However now I have dificulties to plot the  band_tot.dat if somebody could please explain the structure of the output would be fantastic.
> > 
> > 
> > Thank you very much for your attention.
> > Best regards: Rodrigo
> > 
> > 
> > On 11-07-12, Gabriele Penazzi  <gabriele.penazzi at bccms.uni-bremen.de> wrote:
> > > 
> > > 
> > > 
> > > 
> > >   
> > >     
> > >   
> > >   
> > > 
> > > 
> > > 
> > > 
> > >     
> > > On 07/11/2012 09:04 AM, R.Perez Garcia
> > >       wrote:
> > > 
> > >     
> > > 
> > >     
> > > > 
> > > >         
> > > >             
> > > >               
> > > > > 
> > > > >                 Dear Prof. Aradi,
> > > > > 
> > > > >               
> > > > 
> > > >               Dear DFTB+ especialists,
> > > > 
> > > >               
> > > > > 
> > > > >                 
> > > > > 
> > > > >                   
> > > > > 
> > > > >                   
> > > > > 
> > > > >                   I am starting to use DBTB+ and I successfully did so
> > > > >                   until now. 
> > > > > 
> > > > >                   I can not calculate the band structure due to a
> > > > >                   convergence problem.
> > > > > 
> > > > >                   I did follow your tutorial, and after SCC convergence
> > > > >                   (21iterations) and satisfactorily calculate the DOS,
> > > > >                   the BandStructure calculation crashes.
> > > > > 
> > > > >                   
> > > > > 
> > > > >                   I put the necessary files, charges.bin and geom.gen,
> > > > >                   in a new directory  together with the dftb_in.hsd
> > > > >                   (with all the small changes suggested by you, I think
> > > > >                   ) but it can not converged.[cut]
> > > > > 
> > > > >               
> > > > 
> > > >             
> > > > 
> > > >           
> > > > 
> > > >       
> > > 
> > >     
> > > 
> > >     Hi Rodrigo,
> > > 
> > >     
> > > 
> > >     from your input file and output error it looks like you're trying to
> > >     solve the SCC cycle again. This is wrong, you must first calculate
> > >     the charge density with a k-point sampling suitable for that, and
> > >     then import the charge in the band structure calculation WITHOUT
> > >     solving the scc cycle again. Otherwise the code will try to  solve
> > >     it using as a k-point sampling the k lines you specified, that's why
> > >     it doesn't converge.
> > > 
> > >     
> > > 
> > >     Check the online example:
> > > 
> > >     
> > >     http://www.dftb-plus.info/fileadmin/DFTB-Plus/public/recipes/html/basics/bandstruct.html
> > > 
> > >     
> > > 
> > >     In the section "Calculating the band structure" in the input file
> > >     there's a flag "ReadInitialCharges = Yes"
> > > 
> > >     Copy again your charges from the converged calculation and use this
> > >     flag (which gives instruction to the code not to calculate the
> > >     charge again), and probably your problem will be gone.
> > > 
> > >     
> > > 
> > >     Gabriele
> > > 
> > >     
> > > 
> > >     
> > > 
> > >     
> > > 
> > >     
> > > -- 
> > > Dr. Gabriele Penazzi
> > > BCCMS - University of Bremen
> > > 
> > > http://www.bccms.uni-bremen.de/
> > > http://sites.google.com/site/gabrielepenazzi/
> > > 
> > > phone: +49 (0) 421 218 62337(tel:%2B49%20%280%29%20421%20218%2062337)
> > > fax: +49 (0) 421 218 62770(tel:%2B49%20%280%29%20421%20218%2062770)
> > > 
> > >   
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > 
> > 
> > 
> > 
> > _______________________________________________
> > 
> > DFTB-Plus-User mailing list
> > 
> > DFTB-Plus-User at dftb-plus.info
> > 
> > http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user
> > 
> > 
> > 
> 
> 
> 
> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.dftb-plus.info/pipermail/dftb-plus-user/attachments/20120711/777dc419/attachment-0001.html>


More information about the DFTB-Plus-User mailing list