convergence of the electronic wavefunction for metal surface

[ires]

Hi,

maybe someone can help me with this: I'm trying to perform a geometry
optimization of C2H2 on Cu(100), and I can't get the electronic SCF to
convergence without using OT (with OT it's no problem, but since it is
a metallic surface, I thought it's better not to use OT?)
I tried adding MOs and also smearing - here is one of the inputs I
used:

&FORCE_EVAL
METHOD Quickstep
&DFT
BASIS_SET_FILE_NAME /users/depart1/irespond/cp2k/BASIS_MOLOPT
POTENTIAL_FILE_NAME /users/depart1/irespond/cp2k/GTH_POTENTIALS
# RESTART_FILE_NAME ./cuc2h2_pbe-RESTART.wfn
&MGRID
CUTOFF 600
&END MGRID
&QS
EPS_DEFAULT 1.0E-8
&DISTRIBUTION
SKIP_OPTIMIZATION TRUE
&END DISTRIBUTION
&END QS
&SCF
EPS_SCF 1.0E-7
SCF_GUESS ATOMIC
MAX_SCF 500
ADDED_MOS 50
&DIAGONALIZATION T
&END DIAGONALIZATION
&OUTER_SCF
OPTIMIZER DIIS
MAX_SCF 30
&END OUTER_SCF
&SMEAR
METHOD FERMI_DIRAC
WINDOW_SIZE 0.007349862
&END SMEAR
&END SCF
&XC
DENSITY_CUTOFF 1.0e-9
&XC_GRID
XC_SMOOTH_RHO SPLINE3
XC_DERIV SPLINE3_SMOOTH
&END XC_GRID
&XC_FUNCTIONAL
&PBE
&END PBE
&END XC_FUNCTIONAL
&END XC
&END DFT
&SUBSYS
&CELL
ABC 7.3 7.3 18.65
&END CELL
&COORD
Cu 0.0000000000 0.0000000000 0.0000000000
Cu 1.8250000000 1.8250000000 0.0000000000
Cu 0.0000000000 3.6500000000 0.0000000000
Cu 1.8250000000 -1.8250000000 0.0000000000
Cu 3.6500000000 0.0000000000 0.0000000000
Cu -1.8250000000 1.8250000000 0.0000000000
Cu 3.6500000000 3.6500000000 0.0000000000
Cu -1.8250000000 -1.8250000000 0.0000000000
Cu -1.8250000000 0.0000000000 1.8250000000
Cu 3.6500000000 1.8250000000 1.8250000000
Cu -1.8250000000 3.6500000000 1.8250000000
Cu 3.6500000000 -1.8250000000 1.8250000000
Cu 1.8250000000 0.0000000000 1.8250000000
Cu 0.0000000000 1.8250000000 1.8250000000
Cu 1.8250000000 3.6500000000 1.8250000000
Cu 0.0000000000 -1.8250000000 1.8250000000
Cu 0.0027629890 0.1810678691 3.5521582796
Cu 2.0009703441 1.8306590329 3.7419132473
Cu 0.0012714586 3.5522008702 3.5509776811
Cu 1.6331335266 -1.8263840871 3.6506701186
Cu -3.6527301762 -0.2667803211 3.6469179283
Cu -1.9117113169 1.8216433088 3.8247004427
Cu -3.6354479501 4.00316486646 3.6502743039
Cu -1.5520623914 -1.7336304355 3.6519202615
C -3.6540423726 1.2728466111 5.0359160553
C -3.6577740344 2.5275816699 5.0430626704
H -3.6575062980 0.6958416548 5.9483529244
H -3.6560211188 3.0906730999 5.9596030311
&END COORD
&KIND Cu
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PBE-q11
&END KIND
&KIND C
BASIS_SET TZV2P-MOLOPT-GTH
POTENTIAL GTH-PBE-q4
&END KIND
&KIND H
BASIS_SET TZV2P-MOLOPT-GTH
POTENTIAL GTH-PBE-q1
&END KIND
&END SUBSYS
&END FORCE_EVAL
&GLOBAL
PROJECT cuc2h2_pbe
PREFERRED_FFT_LIBRARY FFTSG
RUN_TYPE GEO_OPT
PRINT_LEVEL LOW
&END GLOBAL
&MOTION
&CONSTRAINT
&FIXED_ATOMS
LIST 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
&END FIXED_ATOMS
&END CONSTRAINT
&GEO_OPT
MAX_ITER 300
MAX_FORCE 1.E-7
OPTIMIZER BFGS
&END GEO_OPT
&END MOTION


Any help/tips/comments/suggestions would be greatly appreciated.

[Matthias Krack]

Hi ?,

you request a smearing based on a Fermi-Dirac distribution, but you
didn't specify any electronic temperature for it. Thus you rely on the
default value of 300K for the electronic temperature which might be
too small. Note, the keyword WINDOW_SIZE should be used with the
method ENERGY_WINDOW. This might be at least one issue to improve the
convergence.

Matthias

[Juerg Hutter]

Hi

here is a SCF section we use for Rh(111) surfaces

&SCF
SCF_GUESS RESTART
EPS_SCF 5.0E-7
MAX_SCF 500
ADDED_MOS 500
CHOLESKY INVERSE
&SMEAR ON
METHOD FERMI_DIRAC
ELECTRONIC_TEMPERATURE [K] 300
&END SMEAR
&DIAGONALIZATION
ALGORITHM STANDARD
&END DIAGONALIZATION
&MIXING
METHOD BROYDEN_MIXING
ALPHA 0.1
BETA 1.5
NBROYDEN 8
&END
&PRINT
&RESTART
&EACH
QS_SCF 50
&END
ADD_LAST NUMERIC
&END
&END
&END SCF

best regards

Juerg Hutter

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

> --
>
> You received this message because you are subscribed to the Google Groups "cp2k" group.
> To post to this group, send email to cp...@googlegroups.com.
> To unsubscribe from this group, send email to cp2k+uns...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/cp2k?hl=en.
>
>
>

[ires]

Hi,

thanks for the helpful replies! In the end, the mixing seemed to do
the trick, and the wave function converged nicely.

Inga