[CPMD-list] Si geometry optimization test

[Emre Ersin]

Hi, 
I've changed my input file as follows;

&INFO
 Bulk Si
 GeoMetry Optimization
&END

&CPMD
  ODIIS NO_RESET=50
  5
  OPTIMIZE GEOMETRY XYZ
  CONVERGENCE INITIAL
   1.0d-4
  CONVERGENCE GEOMETRY
   1.0d-3
  CONVERGENCE ORBITALS
   1.0d-7
  CONVERGENCE ADAPT
   0.02
  CONVERGENCE ENERGY
   0.05
  STEEPEST DESCENT CELL
  BFGS
&END

&SYSTEM
  SYMMETRY
  1
  SCALE
  ISOTROPIC CELL
  CELL 
  10.25 1.0 1.0 0.0 0.0 0.0
  CUTOFF
  20.0
&END

&DFT
 FUNCTIONAL GGA
&END

&ATOMS
*SI.USP FORMATTED
   LMAX=p
   64
0.000   0.000   0.000
0.250   0.250   0.000
0.250   0.000   0.250
...


And got this error message;
...
 172  1.140E-03   1.549E-05     -82.186623   -2.278E-05      0.94
 173  9.196E-04   1.910E-05     -82.186606    1.684E-05      0.95
 174  7.912E-04   9.867E-06     -82.186600    5.913E-06      0.94
 175  8.565E-04   9.753E-06     -82.186603   -2.427E-06      0.95
 176  7.203E-04   1.390E-05     -82.186584    1.886E-05      0.94
999
 177  2.087E-04   4.571E-06     -82.186581    2.910E-06      0.95
 STOPGM! STACK OF MAIN CALLS:
 STOPGM! CALL    FORCEDR
 STOPGM! CALL     FORCES
 STOPGM! CALL     STRESS


 PROGRAM STOPS IN SUBROUTINE STRESS| STRESS TENSOR IS BROKEN FOR VANDERBILT
USPP

Is there anything that I can do with my input file to not to get this
message? 
Emre

-----Original Message-----
From: Juerg Hutter [mailto:hutter at pci.unizh.ch] 
Sent: Friday, April 28, 2006 3:15 PM
To: Emre Ersin
Cc: cpmd-list at cpmd.org
Subject: Re: [CPMD-list] Si geometry optimization test

Hi

the box length of silicon in a simple cubic lattice is
10.28 Bohr, not 6 Bohr as in your input.

The FCC structure of silicon in a 8 atom supercell is
not stable, unless you are using k-points.
You have to go to at least a 16 atom supercell to
stabilize the structure in a Gamma point calculation.

I assume you plan to go to other systems and therefore
want to test USPPs. For silicon alone, any norm-conserving
PP would be also ok at 20 Rydberg cutoff.

regards

Juerg Hutter


----------------------------------------------------------
Juerg Hutter                   Phone : ++41 44 635 4491
Physical Chemistry Institute   FAX   : ++41 44 635 6838
University of Zurich           E-mail: hutter at pci.unizh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
----------------------------------------------------------


On Fri, 28 Apr 2006, Emre Ersin wrote:

> Hi,
>
> I am working on geometry optimization of Bulk Si crystal as a beginning
> project.  I am using Vanderbild USPP-735 and cpmd 3.11 (RHEL amd
> x86_64-ifort) . Results in my GEOMETRY.xyz file are completely wrong. What
> can I do about this?
>
> Thanks in advance
>
>
> Result (GEOMETRY.xyz):
> -------------------------------------
>       8
> GEOMETRY FILE / created by CPMD
> Si     -0.095815486591     -0.211103768459     -0.530135268170
> Si      0.399347278859      1.647314818387     -0.616958156691
> Si      1.601381554440     -0.889147011651      1.794392840814
> Si     -0.564859054511      0.740283773580      1.553802768307
> Si      0.311613943292      0.714749168784      0.493894683966
> Si      1.487911875733      2.356276096179      0.300978024842
> Si      1.278324047112      0.698727831026      2.082213155523
> Si     -0.195579241049      2.358727516743      1.058717791456
>
>
> My Input file is;
> ----------------------
>
> &INFO
> Bulk Si
> GeoMetry Optimization
> &END
>
> &CPMD
>  ODIIS NO_RESET=100
>  5
>  HESSIAN UNITY
>  OPTIMIZE GEOMETRY XYZ
>  CONVERGENCE INITIAL
>   1.0d-4
>  CONVERGENCE GEOMETRY
>   1.0d-3
>  CONVERGENCE ORBITALS
>   1.0d-7
>  CONVERGENCE ADAPT
>   0.02
>  CONVERGENCE ENERGY
>   0.05
>  BFGS
> &END
>
> &SYSTEM
>  SYMMETRY
>  1
>  SCALE
>  ISOTROPIC CELL
>  CELL
>  6.0 1.0 1.0 0.0 0.0 0.0
>  CUTOFF
>  20
> &END
>
> &DFT
> FUNCTIONAL GGA
> &END
>
> &ATOMS
> *SI.USP FORMATTED
>   LMAX=p
>   8
> 0       0       0
> 0.5     0.5     0
> 0.5     0       0.5
> 0       0.5     0.5
> 0.25    0.25    0.25
> 0.75    0.75    0.25
> 0.75    0.25    0.75
> 0.25    0.75    0.75
>
> &END
>
> ---
>
> Emre Ersin
>
> _______________________________________________
> CPMD-list mailing list
> CPMD-list at cpmd.org
> http://cpmd.org/mailman/listinfo/cpmd-list
>