[CPMD-list] Atoms are very close?

Erik Santiso eesantis at unity.ncsu.edu
Sat Mar 27 18:07:01 CET 2004


Hi,

Many thanks for your reply! I didn't know these details.

What I've been doing after I got this error message is to do several runs
with the same structure but changing the cell size, i.e. multiply all
coordinates in one direction by a factor to see which one gives the lowest
energy. It's a bit tedious, but since the system is not so large it can be
done. The problem I see is that, without the surface option, it seems I'd
need to have a very large cell in the z direction. If anybody knows of a
less brute-force way to do this please let me know.

Thanks again,

Erik.

-------------------------------------
Vir prudens non contra ventum mingit.

> On Fri, 26 Mar 2004, Erik Santiso wrote:
>
> hello erik,
>
> from looking at your input file, you have some serious problems:
>
> - the surface option only works in combination with symmetry 0.
> - the cell optimization _always_ works in 3 dimensions and
>   _implicitely_ changes the cell symmetry to triclinic (IBRAV=14)
>   unless you set ISOTROPIC CELL. (please have a look at the files
>   sysin.F and setsys.F)
> - for cell optimizations you need the stress tensor, but the stress
>   tensor calculation for vanderbilt pseudopotential is (silently)
>   broken in cpmd v3.7.x. which would explain the error you get.
>
> so the best you can probably do, is run a series of jobs with
> different cell geometries and pick the geometry that seems best.
>
> if i am not mistaken, you have to be careful when comparing the
> energies directly, as the surface energy should also change if
> you change the box along the surface. perhaps somebody else
> would like to comment on that (i'd be interested to learn more
> about how to 'optimize a surface' properly as well).
>
> axel kohlmeyer.
>
> ES>
> ES> Hi,
> ES>
> ES> I'm trying to do a cell optimization on a carbon surface. If I run the
> ES> geometry optimization only, the program runs fine. However, when I add
> ES> STEEPEST DESCENT CELL, I get this error:
> ES>
> ES>  ATOM TYPE=           1  NUM=           1   9.39855555274434
> ES>    8.13938721314203        15.0000000000000
> ES>  ATOM TYPE=           1  NUM=           2   10.7412063475464
> ES>    10.4649264196224        15.0000000000000
> ES>
> ES>
> ES>  PROGRAM STOPS IN SUBROUTINE  SETSYS| ATOMS ARE VERY CLOSE
> ES>
> ES> Even accounting for the PBC in x and y, the atoms are not really very
> ES> close (the distance is roughly a C=C bond distance), unless cpmd is
> ES> somehow changing the cell parameters I'm defining. Am I doing
> something
> ES> wrong? The input I'm using is:
> ES>
> ES> &CPMD
> ES>
> ES>  OPTIMIZE GEOMETRY
> ES>  MEMORY BIG
> ES>  STEEPEST DESCENT CELL
> ES>  CONVERGENCE CELL
> ES>   0.01
> ES>
> ES> &END
> ES>
> ES> &SYSTEM
> ES>  POISSON SOLVER MORTENSEN
> ES>  SURFACE
> ES>  SYMMETRY
> ES>   8
>
> the surface option only works in combination with symmetry 0.
>
> the cell optimization
>
> ES>  CELL ABSOLUTE
> ES>   24.1677142816811 18.6043136308668 30.0 0.0 0.0 0.0
> ES>  CUTOFF
> ES>   35
> ES>
> ES> &END
> ES>
> ES> &ATOMS
> ES>
> ES> *C_BLYP_wcc.uspp FORMATTED NEWF NLCC
> ES>  LMAX=P
> ES>  48
> ES>   1.889726625715786E-002  1.889726625715786E-002
> 0.000000000000000E+000
> ES>    1.36154806105919        2.34443647273750
> 0.000000000000000E+000
> ES>    1.36154806105919       -2.30664193946729
> 0.000000000000000E+000
> ES>   -2.66640432221306       1.889726625715786E-002
> 0.000000000000000E+000
> ES>   1.889724868269989E-002   4.66997567883990
> 0.000000000000000E+000
> ES>    4.04684964990735        2.34443647273750
> 0.000000000000000E+000
> ES>    4.04684964990735       -2.30664193946729
> 0.000000000000000E+000
> ES>   1.889724868269989E-002  -4.63218114556969
> 0.000000000000000E+000
> ES>   -4.00905511663714        2.34443647273750
> 0.000000000000000E+000
> ES>   -4.00905511663714       -2.30664193946729
> 0.000000000000000E+000
> ES>    1.36154806105919        6.99551488494229
> 0.000000000000000E+000
> ES>   -2.66640432221306        4.66997567883990
> 0.000000000000000E+000
> ES>    5.38950044433143        4.66997567883990
> 0.000000000000000E+000
> ES>    5.38950044433143       1.889724868269989E-002
> 0.000000000000000E+000
> ES>    5.38950044433143       -4.63218114556969
> 0.000000000000000E+000
> ES>   -2.66640432221306       -4.63218114556969
> 0.000000000000000E+000
> ES>    1.36154806105919       -6.95772035167209
> 0.000000000000000E+000
> ES>   -6.69435670548531        2.34443647273750
> 0.000000000000000E+000
> ES>   -6.69435670548531       -2.30664193946729
> 0.000000000000000E+000
> ES>   1.889724868269989E-002   9.32105409104469
> 0.000000000000000E+000
> ES>    4.04684964990735        6.99551488494229
> 0.000000000000000E+000
> ES>   -4.00905511663714        6.99551488494229
> 0.000000000000000E+000
> ES>    8.07480203317959        4.66997567883990
> 0.000000000000000E+000
> ES>    8.07480203317959       1.889724868269989E-002
> 0.000000000000000E+000
> ES>    8.07480203317959       -4.63218114556969
> 0.000000000000000E+000
> ES>    4.04684964990735       -6.95772035167209
> 0.000000000000000E+000
> ES>   -4.00905511663714       -6.95772035167209
> 0.000000000000000E+000
> ES>   -8.03700749990939        4.66997567883990
> 0.000000000000000E+000
> ES>   -8.03700749990939       1.889724868269989E-002
> 0.000000000000000E+000
> ES>   -8.03700749990939       -4.63218114556969
> 0.000000000000000E+000
> ES>   -2.66640432221306        9.32105409104469
> 0.000000000000000E+000
> ES>    5.38950044433143        9.32105409104469
> 0.000000000000000E+000
> ES>   -6.69435670548531        6.99551488494229
> 0.000000000000000E+000
> ES>    9.41745282760368        6.99551488494229
> 0.000000000000000E+000
> ES>    9.41745282760368        2.34443647273750
> 0.000000000000000E+000
> ES>    9.41745282760368       -2.30664193946729
> 0.000000000000000E+000
> ES>    9.41745282760368       -6.95772035167209
> 0.000000000000000E+000
> ES>   -6.69435670548531       -6.95772035167209
> 0.000000000000000E+000
> ES>    8.07480203317959        9.32105409104469
> 0.000000000000000E+000
> ES>    12.1027544164518        6.99551488494229
> 0.000000000000000E+000
> ES>    12.1027544164518        2.34443647273750
> 0.000000000000000E+000
> ES>    12.1027544164518       -2.30664193946729
> 0.000000000000000E+000
> ES>    12.1027544164518       -6.95772035167209
> 0.000000000000000E+000
> ES>   -8.03700749990939        9.32105409104469
> 0.000000000000000E+000
> ES>    13.4454052108759       -4.63218114556969
> 0.000000000000000E+000
> ES>    13.4454052108759       1.889726663510319E-002
> 0.000000000000000E+000
> ES>    13.4454052108759        4.66997567883990
> 0.000000000000000E+000
> ES>    13.4454052108759        9.32105409104469
> 0.000000000000000E+000
> ES>
> ES> &END
> ES>
> ES> &DFT
> ES>
> ES>   FUNCTIONAL BLYP
> ES>   GC-CUTOFF
> ES>    5.D-5
> ES>
> ES> &END
> ES>
> ES> Any help would be greatly appreciated. Thanks!
> ES>
> ES> Erik.
> ES>
> ES> -------------------------------------
> ES> Vir prudens non contra ventum mingit.
> ES> _______________________________________________
> ES> CPMD-list mailing list
> ES> CPMD-list at cpmd.org
> ES> http://www.cpmd.org/mailman/listinfo/cpmd-list
> ES>
> ES>
>
> --
>
>
> =======================================================================
> Dr. Axel Kohlmeyer                        e-mail: axel.kohlmeyer at rub.de
> Lehrstuhl fuer Theoretische Chemie          Phone: ++49 (0)234/32-26673
> Ruhr-Universitaet Bochum - NC 03/53         Fax:   ++49 (0)234/32-14045
> D-44780 Bochum  http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/
> =======================================================================
>
>
>






More information about the CPMD-list mailing list