The ABINIT Users Mailing List ( CLOSED )

[matthieu verstraete <matthieu.jean.verstraete@gmail.com>]

abinit always symmetrizes forces and stresses, so the space group is preserved (this is what happens for ionmov 1). However, with Langevin there are extra stochastic terms, which do not obey symmetry, and break the space group. This is probably what you want, but it means, as Pierre-Matthieu notes, that you have to ignore symmetries from the beginning (nsym 1 or move atoms slightly to break the symmetries). Abinit is very rigorous about finding and preserving symmetries.

We might put an extra warning in the code, in case of Langevin dynamics which we know will break symmetries, or have the code skip the symmetry testing in that case... But the rigorous thing to do is to remove symmetries from the start (or keep some, if the dynamics will not break them).

Matthieu

On Fri, Feb 27, 2009 at 11:42 AM, Anglade Pierre-Matthieu <anglade@gmail.com> wrote:

Hi,

I wonder if this problem could be solved by enforcing "nsym 1" in your system ?

regards

PMA

On Fri, Feb 27, 2009 at 11:18 AM, Quan Phung Manh <manhquank9@gmail.com> wrote:
> I'm learning CO on TiO2 surface using Molecular dynamics. This is the input
> file
>
> #Definition of the k-point grids
> kptopt 1
> nshiftk 2
> shiftk  0.5 0.0 0.0
>         0.0 0.5 0.0
> ngkpt  2 2 1
>
> #Definition of the unit cell
> acell 5.5045254737E+00  1.2085958617E+01  3.9054040855E+01
>
> #Definition of the atom types
> ntypat 5
> znucl 22 8 1 6 8
>
> #Definition of the atoms
> natom 27
> typat 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 4 5
> xangst
> 0.000000000     3.433979661     1.702548614
> 0.000000000    -0.030909893     4.965652343
> 0.000000000     3.183736244     8.301964739
> 0.000000000    -0.007453315    11.238648984
> 1.456434713     0.249962141     1.826819536
> 1.456434713     3.160658047     4.997001953
> 1.456434713    -0.015133698     8.106539672
> 1.456434713     3.191007391    11.602913960
> 0.000000000     3.134088248     3.712077931
> 0.000000000    -0.036979110     6.917159950
> 0.000000000     3.187105620    10.148049014
> 0.000000000     0.085095967     3.023282974
> 0.000000000     3.185355376     6.258969546
> 0.000000000    -0.016913131     9.441795449
> 0.000000000     3.189623132    12.679519266
> 1.456434713     4.628213937     1.641901518
> 1.456434713     1.210599710     5.011844303
> 1.456434713     4.461364915     8.201911335
> 1.456434713     1.198244350    11.547258895
> 1.456434713     2.045849552     1.627326844
> 1.456434713     5.122602774     4.938437347
> 1.456434713     1.894988115     8.197087652
> 1.456434713     5.182999613    11.552484499
> 0.000000000     3.197805952     0.000000000
> 2.912869425     0.000000000     0.700956091
> 0.000000000    -0.007453315    13.738648984
> 0.000000000    -0.007453315    14.866648984
>
> natfix 25
> iatfix 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
>
> #Definition of the planewave basis set
> ecut  40.0
>
> #Definition of the SCF procedure
> nstep 200              # Maximal number of SCF cycles
> tolmxf  5.0d-4
> toldff  5.0d-5
>
> #Optimization of the lattice parameters
> ionmov  9              # Langevin molecular dynamics
> ntime  20
> mditemp 100
> iprcel 45
>
> occopt 3
> tsmear 9.5d-5
>
> And I have many warnings in MOLDYN STEP NUMBER    1. For example
>  symatm : WARNING -
>   Trouble finding symmetrically equivalent atoms
>   Applying inv of symm number  2 to atom number   1  of typat  1
>   gives tratom= -1.0111E-03  5.3775E-01  8.2374E-02.
>   This is further away from every atom in crystal than the allowed
> tolerance.
>   The inverse symmetry matrix is -1  0  0
>                                   0  1  0
>                                   0  0  1
>   and the nonsymmorphic transl. tnons =    0.0000000    0.0000000
> 0.0000000
>   The nearest coordinate differs by -2.022E-03  0.000E+00  0.000E+00
>   for indsym(nearest atom)=    1
>
>   This indicates that when symatm attempts to find atoms symmetrically
>   related to a given atom, the nearest candidate is further away than some
>   tolerance.  Should check atomic coordinates and symmetry group input data.
>
> and an error
>  symatm : ERROR -
>   Largest error (above) is so large that either input  atomic coordinates
> (xred)
>   are wrong or space group symmetry data is wrong.
>   Action : correct your input file.
>
>  leave_new : decision taken to exit ...
>
> I choose ionmov = 1 and have no error.
>
> Is there anything wrong with the input file?
>
> Best regards
> --
> Quan Phung Manh
> Khoa Hoá học - ĐH Khoa học tự nhiên - 19 Lê Thánh Tông Hà Nội
> ĐT: 01663 922 302
>

--
Pierre-Matthieu Anglade

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Matthieu Verstraete

European Theoretical Spectroscopy Facility (ETSF) 
Dpto. Fisica de Materiales, 
U. del Pais Vasco,            
Centro Joxe Mari Korta, Av. de Tolosa, 72,   Phone: +34-943018393
E-20018 Donostia-San Sebastian, Spain        Fax  : +34-943018390

Mail : matthieu.jean.verstraete@gmail.com
http://www-users.york.ac.uk/~mjv500