[Wien] structural minimization

[Peter Blaha]

Hard to say!

Actually, not the second, but the 3rd one is wrong!
The second has still proper symmetry with a shift -dx,-dx,+dx as it should
be with these forces.

The 3rd structure, the x-coordinate gets wrong.

I could see in outputM that  for this step
   pp = 
      0.00000000      0.00000000      0.00000000
      0.00645000      0.00645000      0.00645000
     -0.00645000     -0.00645000      0.00645000
     -0.00645000      0.00645000     -0.00645000
      0.00645000     -0.00645000     -0.00645000
      0.00645000      0.00645000     -0.00645000
      0.00645000     -0.00645000      0.00645000
     -0.00645000      0.00645000      0.00645000
     -0.00645000     -0.00645000     -0.00645000
     -0.00373400     -0.00175100      0.00175100
This last line can't be ok, since the x and y part must be the
same (and it was the step before) and this results in wrong positions.

I've not used NEWT for the last 6 month, since it is no longer recommended.
If you still have an earlier version of mini, you could cross check if this
is a new error introduced by the PORT changes, or was also present before.

Anyway, I'd recommend using PORT.

For further debugging I would need the scf_mini file to have all forces
and to redo the mini steps.

Regards

> The structure was constructed on a usual bcc-type lattice: one impurity in a 
> 32-atoms cell in FCC symmetry (the Fe31Sn_1.struct file is attached). The 
> first step was normal, but the second one of the third atom looks strange for 
> me:
> Fe31Sn_1.struct:ATOM  -3: X=0.12500000 Y=0.12500000 Z=0.37500000
> Fe31Sn_2.struct:ATOM  -3: X=0.12490905 Y=0.12490905 Z=0.37509095
> Fe31Sn_3.struct:ATOM  -3: X=0.12473780 Y=0.12482875 Z=0.37517125
> The forces were not unsymmetrical in the second iteration:
> :FGL003:   3.ATOM                        -3.966         -3.966          3.966
> :FGL003:   3.ATOM                        -3.502         -3.502          3.502
> :FGL003:   3.ATOM                        -1.476         -3.910          3.909
> Why the program took nonequal shifts and distorted the symmetry? After this 
> step the nn program gives a change of symmetry and new nonequivalent atoms 
> appeared.
> I am attaching the outputM (there are 5 steps already) and Fe31Sn_3.struct
> files. The version of WIEN is of 6 Oct.2004.
> 
> Best regards,
>   Lyudmila Dobysheva 
> ------------------------------------------------------------------
> Phys.-Techn. Institute of       | Tel.(home):   7 (3412) 442118
> Ural Br. of Russian Ac. of Sci. | Tel.(office): 7 (3412) 218988
> 426001 Izhevsk, ul.Kirova 132   | Fax:          7 (3412) 250614
> RUSSIA                          | E-mail: lyu at otf.fti.udmurtia.su
> ------------------------------------------------------------------
> http://fti.udm.ru/ltt/personals/dobysh.htm
> ------------------------------------------------------------------
> 


                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------