[CPMD-list] Geometry Optimization of O on Pt slab

[Axel Kohlmeyer]

>>> "XT" == Xudong Tang <knba713 at hotmail.com> writes:

dear xudong,

you can try to increase the broyden mixing. 
e.g. use 0.5, or 0.3 instead of 0.15. 

this has helped me in some (non-metallic though) cases.

axel.

XT> hi, CPMD users,

XT> I'm tring to use CPMD simulating the adsorbed position of Oxygen atom on Pt 
XT> slab. But with local spin density approximation, the force initialization 
XT> process is very slow, which needs like 600+ step to get converged. Does 
XT> anyone here have the experience to accelerate this iteration?

XT> Thanks.

XT> Xudong

XT> Below is the input file for Pt(100) surface:
XT> ------------------------------------------------------------------------
XT> &CPMD
XT>     OPTIMIZE GEOMETRY
XT>     LSD
XT>     UNIT HESSIAN
XT>     BFGS
XT>     FREE ENERGY FUNCTIONAL
XT>     LANCZOS DIAGONALISATION
XT>     LANCZOS PARAMETERS
XT>       1  6 15   1.D-18
XT>     TROTTER FACTOR
XT>       0.001
XT>     BOGOLIUBOV CORRECTION OFF
XT>     GRAM-SCHMIDT ORTHOGONALISATION
XT>     CONVERGENCE
XT>       5.D-4   1.D-3
XT>     MAXSTEP
XT>       5000
XT>     BROYDEN MIXING
XT>       0.15 200   0.01  0   10
XT>     ALEXANDER MIXING
XT>       1.1
XT>     ELECTRON TEMPERATURE
XT>       1000.
XT>     COMPRESS WRITE32
XT>     STRUCTURE BONDS

XT>     ANDERSON MIXING
XT>       0.03
XT> &END



--

=======================================================================
Axel Kohlmeyer       e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.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/
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.