The ABINIT Users Mailing List ( CLOSED )

[Masayoshi Mikami <mmikami@rc.m-kagaku.co.jp>]

Dear Nguyen Ha,

To add just one more comment to Xavier's,
I believe that "ecut 15" should be too small,
if you rely on ordinary pseudopotentials.
(e.g. ecut 30-35 Hartree for oxygen pseudopotential
 based on Troullier-Martins. They say OPIUM can
 produce softer PP: http://opium.sourceforge.net/).
Thus, such information should be also provided.

Kindly peruse the Netiquette, before you post:
http://www.abinit.org/community/?text=netiquette

Good luck,
Masayoshi

P.S. I did not check the cell, but it will be a good idea
to visualize your cell before/after optimizing crystal cell.

On 2006/08/08, at 12:48, Xavier Gonze wrote:

> Dear nh nh,
>
> I see at least one problem : you should use another stopping  
> criterion than toldfe 1.0d-6 .
> Use e.g. tolvrs, with a value that might be 1.0d-12 .
>
> Hard to tell more without the output file.
> If you have other problems, please set ntime to 10 nstep to 20  
> (this is to generate
> an output file of reasonable size) and send your output file.
>
> MgO being an insulator, you
> should succeed to use the minimal set of bands, so either suppress  
> again nband, or
> set it to the minimal number needed for your pseudopotentials.
>
> Xavier
>
> On 08 Aug 2006, at 03:15, nh nh wrote:
>
> >    Dear Abinit Users, I can't opt MgO cell. Although I have used  
> > both primitive form and conventional form; I raise nband to 50!.  
> > For example, here is my input (in primitive form):
> >
> >
> > #Optimization of the lattice parameters
> > optcell 2
> > ionmov 3
> > ntime 300
> > dilatmx 1.5
> > ecutsm 0.5
> > getwfk -1
> > prtgeo 1
> > nband 50
> > #spgroup 225
> > #brvltt -1
> > #Definition of the unit cell
> > acell 2.9778 2.9778 2.9778 angstrom
> > angdeg 60 60 60
> > #chkprim 0
> > #Definition of the atom types
> > ntypat 2
> > znucl 12 8
> > #Definition of the atoms
> > natom 2
> > typat 1 2
> > xred
> > 0.0 0.0 0.0
> > 0.500000 0.500000 0.500000
> > ecut 15.0
> > nstep 300
> >  toldfe 1.0d-6
> > diemac 2.0
> >
> >
> > ****
> > At the end, I see:
> > scfcge : ERROR -
> >  Potential-based CG line minimization not converged after 13  
> > restarts.
> >  Action : read the eventual warnings about lack of convergence.
> >  Some might be relevant. Otherwise, raise nband.
> > leave_new : decision taken to exit ...
> >
> >          Could you help me for this! Thank you very much!
> >   Nguyen Ha
> > Send instant messages to your online friends http:// 
> > uk.messenger.yahoo.com