[CP2K-user] [CP2K:16946] Re: Cholesky decomposition failed for oxide surface geometry optimization

[Sam Broderick]

OT may be running into trouble with Cu. For metals, it is not recommended. 
However, this is a compound...
I ran into a similar issue with an Au cluster and addressed it by using 
Davidson diagonalization.


KR
Sam
On Thursday, July 23, 2020 at 2:14:07 AM UTC+2 nisa.ulum... at gmail.com wrote:

> Hi,
>
> I am quite new to CP2K and I do not know why I am failing to complete an 
> scf cycle for a Cu2O(111) surface. 
> The error I am getting is :
>
> Number of orbital functions:                                              
>  4536
>  Number of independent orbital functions:                                  
>  4536
>
>  Extrapolation method: initial_guess
>
>
>  *******************************************************************************
>  *   ___                                                                  
>      *
>  *  /   \                                                                  
>     *
>  * [ABORT]                                                                
>      *
>  *  \___/    Cholesky decompose failed: the matrix is not positive 
> definite or *
>  *    |                              ill-conditioned.                      
>     *
>  *  O/|                                                                    
>     *
>  * /| |                                                                    
>     *
>  * / \                                                  
> fm/cp_fm_cholesky.F:94 *
>
>  *******************************************************************************
>
>
>  ===== Routine Calling Stack =====
>
>             8 cp_fm_cholesky_decompose
>             7 make_basis_sm
>             6 calculate_first_density_matrix
>             5 scf_env_initial_rho_setup
>             4 init_scf_run
>             3 qs_energies
>             2 qs_forces
>             1 CP2K
>
>
> And my input file is:
>
>> &FORCE_EVAL
>>   METHOD Quickstep
>>   &DFT
>>     UKS T
>>     !PLUS_U_METHOD MULLIKEN  #_CHARGES
>>     MULTIPLICITY  1
>>     !CHARGE 1
>>     SURFACE_DIPOLE_CORRECTION
>>     SURF_DIP_DIR Z
>>     POTENTIAL_FILE_NAME 
>> /opt/apps/cp2k/6.1/gcc/7.3.0/openmpi/3.0.0/cp2k-6.1.0/data/POTENTIAL
>>     BASIS_SET_FILE_NAME 
>> /opt/apps/cp2k/6.1/gcc/7.3.0/openmpi/3.0.0/cp2k-6.1.0/data/BASIS_MOLOPT
>>     WFN_RESTART_FILE_NAME nu-RESTART.wfn
>>     &MGRID
>>       CUTOFF 400
>>       NGRIDS 4
>>       &INTERPOLATOR
>>         !KIND SPLINE_3
>>       &END INTERPOLATOR
>>     &END MGRID
>>     &QS
>>       EPS_DEFAULT 1.0E-10
>>       EXTRAPOLATION PS
>>       !EXTRAPOLATION_ORDER 2
>>       EXTRAPOLATION_ORDER 2
>>       EPS_PGF_ORB 1E-10
>>     &END QS
>>     &SCF
>>       !CHOLESKY OFF
>>       SCF_GUESS RANDOM
>>      ! SCF_GUESS RESTART
>>       EPS_SCF 1.0e-6
>>       MAX_SCF 400
>>       &OUTER_SCF
>>          OPTIMIZER SD
>>          EPS_SCF 1.0E-6
>>          MAX_SCF 10
>>          STEP_SIZE 0.1
>>       &END
>>       &OT T
>>          !CHOLESKY OFF
>>          MINIMIZER CG
>>          PRECONDITIONER FULL_SINGLE
>>          LINESEARCH 3PNT
>>       &END
>>       &PRINT
>>         !&RESTART
>>         !   FILENAME =RESTART
>>         !&END
>>       &END PRINT
>>     &END SCF
>>     &XC
>>      &XC_GRID
>>       XC_SMOOTH_RHO NN50
>>      &END XC_GRID
>>      &XC_FUNCTIONAL PBE
>>       &PBE
>>            PARAMETRIZATION PBESOL #optimized for solids and surfaces
>>       &END PBE
>>      &END XC_FUNCTIONAL
>>       !&vdW_POTENTIAL
>>         !DISPERSION_FUNCTIONAL PAIR_POTENTIAL
>>         !&PAIR_POTENTIAL
>>            ! TYPE DFTD3
>>            ! REFERENCE_FUNCTIONAL PBE
>>            ! PARAMETER_FILE_NAME /global/homes/g/gbcoll/dftd3.dat
>>            ! R_CUTOFF [angstrom] 15.0
>>            ! VERBOSE_OUTPUT T
>>         !&END PAIR_POTENTIAL
>>       !&END vdW_POTENTIAL
>>     &END XC
>>     &POISSON
>>       &EWALD
>>         EWALD_ACCURACY 1.00e-06
>>       &END EWALD
>>       POISSON_SOLVER ANALYTIC
>>       PERIODIC XY
>>     &END POISSON
>>     &PRINT
>>     &END PRINT
>>   &END DFT
>>   &SUBSYS
>>     &CELL
>>       A  18.3111991882         0.0000000000         0.0000000000
>>       B  -9.1555995941        15.8579636708 <(857)%20963-6708>        
>>  0.0000000000
>>       C  0.0000000000         0.0000000000        28.7215003967
>>       PERIODIC XY
>>     &END CELL
>>     &COORD
>>       @include 'supercell.vasp.xyz'
>>     &END COORD
>>     &KIND Cu
>>       BASIS_SET DZVP-MOLOPT-SR-GTH-q11
>>       POTENTIAL GTH-PBE-q11
>>       !&DFT_PLUS_U  T
>>        ! L               2
>>        ! U_MINUS_J  [eV] 8.0
>>       !&END DFT_PLUS_U
>>       !MASS 8.0
>>     &END KIND
>>     &KIND O
>>       BASIS_SET DZVP-MOLOPT-SR-GTH
>>       POTENTIAL GTH-PBE
>>       !MASS 8.0
>>     &END KIND
>>   &END SUBSYS
>> &END FORCE_EVAL
>> &GLOBAL
>>   PROJECT nu
>>   RUN_TYPE ENERGY_FORCE
>>   PRINT_LEVEL LOW
>>   WALLTIME 01:00:00
>> &END GLOBAL
>
>
>
>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/dbdc7e02-8061-47dd-8e4e-269f0f4806ddn%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20220511/06e852c4/attachment-0001.htm>