Error when running geo_opt
Axel
akoh... at gmail.com
Fri Aug 7 23:57:35 UTC 2009
On Aug 7, 4:53 pm, Eric Shamay <eric.... at gmail.com> wrote:
> Toe,
>
> Thanks for looking into this. The geometry is very wrong - it's only a
> prelim guess of the perfluorodecyl silane that I was hoping to relax
> and minimize using cp2k. I will try to find a better starting geometry
> to work with than what I have and see if that fixes it all up.
why don't you just relax it with a minimal classical force field?
just using parameters as if all heavy atoms were carbons should
do the trick. then you have better starting geometry without overlap
of the cores, which is probably the reason for the failure you are
seeing with QS.
axel.
>
> Thanks!
> ~Eric
>
> On Aug 7, 12:09 pm, Teodoro Laino <teodor... at gmail.com> wrote:
>
> > Hi Eric,
>
> > this error had nothing to do with GEO_OPT.
>
> > The geometry of your system looks a bit weird.. are you sure it is ok
> > (see the nice arrow in the pic ;-) )?!
>
> > Anyway, the error you get is the effect of some linear dependence at the
> > basis set level (or geometry totally wrong).
> > If the geometry is correct, try to use the MOLOPT basis set. It should
> > behave better for these kind of problems (possibly even with
> > the arrow one ;-) )
>
> > Teo
>
> > Eric Shamay wrote:
> > > Hello cp2k,
>
> > > I'm running into trouble when trying to do a GEO_OPT run. I'm posting
> > > the input file and the tail of the error produced down below. If
> > > anyone has run into this type of problem before I'd appreciate your
> > > insight.
>
> > > ****************************************** input file
> > > **************************************
>
> > > &FORCE_EVAL
> > > METHOD Quickstep
> > > &DFT
> > > EXCITATIONS TDDFPT
> > > LSD
> > > &MGRID
> > > CUTOFF 150
> > > NGRIDS 4
> > > REL_CUTOFF 40
> > > &END MGRID
> > > &QS
> > > &END QS
> > > &SCF
> > > &OUTER_SCF
> > > EPS_SCF 3.0E-5
> > > &END OUTER_SCF
> > > EPS_SCF 1.0E-5
> > > MAX_SCF 200
> > > &MIXING
> > > METHOD DIRECT_P_MIXING
> > > ALPHA 0.2
> > > &END
> > > SCF_GUESS ATOMIC
> > > &OT T
> > > MINIMIZER CG
> > > PRECONDITIONER FULL_ALL
> > > ENERGY_GAP 0.500 # make larger if poor initial guess on the
> > > preconditioner
> > > &END OT
> > > &END SCF
> > > &XC
> > > &XC_GRID
> > > XC_DERIV SPLINE2_SMOOTH
> > > XC_SMOOTH_RHO NN10
> > > &END XC_GRID
> > > &XC_FUNCTIONAL BLYP
> > > &PADE
> > > &END PADE
> > > &END XC_FUNCTIONAL
> > > &END XC
> > > &TDDFPT
> > > INVERT_S Y
> > > KERNEL Y
> > > MAX_KV 60
> > > NEV 3
> > > OE_CORR NONE
> > > PRECOND N
> > > REORTHO 2
> > > RES_ETYPE SINGLETS
> > > RESTARTS 4
> > > &END TDDFPT
> > > BASIS_SET_FILE_NAME /home/eric/cp2k/tests/QS/GTH_BASIS_SETS
> > > POTENTIAL_FILE_NAME /home/eric/cp2k/tests/QS/POTENTIAL
> > > &END DFT
> > > &SUBSYS
> > > &CELL
> > > ABC 15.0 10.0 10.0
> > > PERIODIC XYZ
> > > &END CELL
> > > &COORD
> > > C 1.344000 0.000000 0.000000
> > > F 0.000000 0.000000 0.000000
> > > F 1.344000 1.344000 0.000000
> > > F 1.344000 -1.344000 0.000000
> > > C 2.688000 0.000000 0.000000
> > > F 2.688000 1.344000 0.000000
> > > F 2.688000 -1.344000 0.000000
> > > C 4.032000 0.000000 0.000000
> > > F 4.032000 1.344000 0.000000
> > > F 4.032000 -1.344000 0.000000
> > > C 5.376000 0.000000 0.000000
> > > F 5.376000 1.344000 0.000000
> > > F 5.376000 -1.344000 0.000000
> > > C 6.720000 0.000000 0.000000
> > > F 6.720000 1.344000 0.000000
> > > F 6.720000 -1.344000 0.000000
> > > C 8.064000 0.000000 0.000000
> > > F 8.064000 1.344000 0.000000
> > > F 8.064000 -1.344000 0.000000
> > > C 9.408000 0.000000 0.000000
> > > F 9.408000 1.344000 0.000000
> > > F 9.408000 -1.344000 0.000000
> > > C 10.752000 0.000000 0.000000
> > > F 10.752000 1.344000 0.000000
> > > F 10.752000 -1.344000 0.000000
> > > C 12.096000 0.000000 0.000000
> > > H 12.096000 1.344000 0.000000
> > > H 12.096000 -1.344000 0.000000
> > > C 13.440000 0.000000 0.000000
> > > H 13.440000 1.344000 0.000000
> > > H 13.440000 -1.344000 0.000000
> > > Si 14.500000 0.000000 0.000000
> > > &END COORD
> > > &KIND H
> > > BASIS_SET TZV2P-GTH
> > > POTENTIAL GTH-BLYP-q1
> > > &END KIND
> > > &KIND C
> > > BASIS_SET TZV2P-GTH
> > > POTENTIAL GTH-BLYP-q4
> > > MASS 2.0
> > > &END KIND
> > > &KIND F
> > > BASIS_SET TZV2P-GTH
> > > POTENTIAL GTH-BLYP-q7
> > > &END KIND
> > > &KIND Si
> > > BASIS_SET TZV2P-GTH
> > > POTENTIAL GTH-BLYP-q4
> > > &END KIND
> > > &END SUBSYS
> > > &END FORCE_EVAL
> > > &GLOBAL
> > > PROJECT_NAME pds-geo-opt
> > > RUN_TYPE GEO_OPT
> > > PRINT_LEVEL LOW
> > > PREFERRED_FFT_LIBRARY FFTW
> > > &END GLOBAL
>
> > > *******************************************************************************************
> > > ******************************************* error output
> > > *********************************
>
> > > SCF PARAMETERS Density
> > > guess: ATOMIC
>
> > > --------------------------------------------------------
>
> > > max_scf: 200
>
> > > max_scf_history: 0
>
> > > max_diis: 4
>
> > > --------------------------------------------------------
>
> > > eps_scf: 1.00E-05
>
> > > eps_scf_history: 0.00E+00
>
> > > eps_diis: 1.00E-01
>
> > > eps_eigval: 1.00E-05
>
> > > --------------------------------------------------------
> > > level_shift
> > > [a.u.]: 0.00
>
> > > --------------------------------------------------------
> > > Outer loop SCF in use
> > > No variables optimised in outer loop
>
> > > eps_scf 3.00E-05
>
> > > max_scf 50
> > > No outer loop optimization
>
> > > step_size 5.00E-01
>
> > > 2D_MC| Monte Carlo annealing to optimize the distribution_2d
> > > 2D_MC| Number of
> > > moves 400000
> > > 2D_MC| Number of annealing
> > > cycles 10
> > > 2D_MC| Number of reduction steps per
> > > cycle 5
> > > 2D_MC| Reduction factor per step
> > > 0.8000000000
> > > 2D_MC| Termination tolerance
> > > 0.0200000000
> > > 2D_MC| Maximum temperature
> > > 0.5000000000
> > > 2D_MC| Swap probability
> > > 0.9000000000
> > > 2D_MC| Number of processor
> > > rows 1
> > > 2D_MC| Number of processor
> > > cols 1
> > > 2D_MC| Number of
> > > elements 32
> > > 2D_MC| What do we minimize SMALLEST
> > > MAXIMUM
> > > 2D_MC| Cost of optimal
> > > distribution 219490
> > > 2D_MC| Cost of found
> > > distribution 219490
> > > 2D_MC| Difference in
> > > percent 0
>
> > > *******************************************************************************
> > > *** STARTING GEOMETRY
> > > OPTIMIZATION ***
> > > ***
> > > BFGS ***
>
> > > *******************************************************************************
>
> > > Spin 1
>
> > > Number of electrons: 84
> > > Number of occupied orbitals: 84
> > > Number of molecular orbitals: 84
>
> > > Spin 2
>
> > > Number of electrons: 83
> > > Number of occupied orbitals: 83
> > > Number of molecular orbitals: 83
>
> > > Number of orbital functions: 652
> > > Number of independent orbital functions: 652
>
> ...
>
> read more »
More information about the CP2K-user
mailing list