[CPMD-list] LSD/KPOINTS and Fermi energy/FEMD

[Eung-Gun Kim]

Dear Dr. Seitsonen,

Thanks very much for responding to my message.

  1) I must have underestimated the memory requirement of an LSD calculation.
     It works fine now.  The memory error was indeed caused by a failure to 
     allocate memory.
     
  2) The chemical potential becomes equal to the Fermi energy only when T = 0.
     Does this mean that one needs another FEMD calculation at T = 0 (or by
     turning off FEMD), possibly with more k points, to get the Fermi energy?

Regards,

E Kim

Quoting Ari P Seitsonen <Ari.P.Seitsonen at iki.fi>:

> 
> Dear Eung-Gun Kim,
> 
>   1) I don't manage to reproduce your problem, could you please send us at 
> least the output with the complete error message, or better, the input and 
> the pseudo potentials.
> 
>   2) Look for a line like
>  CHEMICAL POTENTIAL =                             2.4181249998 EV
> when the Kohn-Sham eigenvalues are printed out.
> 
>     Greetings,
> 
>        apsi
> 
> -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
>  Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/
>  Tel +41 1 635 44 97 / Fax +41 1 635 68 38 / GSM +41 79 719 09 35      
>  Anschrift: Physikalisch Chemisches Institut (PCI), Universit[e4]t Z[fc]rich
> (UniZh)
>  Indirizzo: Winterthurerstra[df]e 190, CH-8057 Z[fc]rich
>  Address:   Schweiz / Svizzera / Suisse / Svizra / Switzerland
> 
> On Mon, 3 May 2004, Eung-Gun Kim wrote:
> 
> > Dear CPMD List Subscribers,
> > 
> > While running calculations on an organic crystal and a metal slab with
> > CPMD-3.7.2, I have encountered the following two problems:
> > 
> >   1) When I combine LSD with KPOINTS, the program stops with
> memory-related
> > errors such as "PROGRAM STOPS IN SUBROUTINE MEMORY| ALLOCATION FAILED
> (DFNL)
> > [PROC=   0] p0_24396:  p4_error: : 999".  Reducing the system size doesn't
> help.
> > 
> >   2) How do you obtain the Fermi energy from a FEMD calculation?
> > 
> > Thank you very much for reading.
> > 
> > Regards,
> > 
> > E Kim
> > 
> > 
> > _______________________________________________
> > CPMD-list mailing list
> > CPMD-list at cpmd.org
> > http://www.cpmd.org/mailman/listinfo/cpmd-list
> > 
>