[CP2K-user] [CP2K:13897] Need advice on bulk metal system calculations

Brendan Smith bsmi... at buffalo.edu
Tue Sep 15 20:11:54 UTC 2020


Hi Matthias,

I have a separate question regarding the 2x2x2 super cell. What if I make a 2x2x2 super cell manually and use the super cell atomic coordinates and cell parameters in the input file instead of the multiple unit cell 2x2x2 option. Would it be the same as using the single conventional unit cell but with the multiple unit cell 2x2x2 option turned off?

Kind regards,
Brendan Smith
SUNY Buffalo

> On Sep 15, 2020, at 12:40 PM, Krack Matthias (PSI) <matthi... at psi.ch> wrote:
> 
> 
> Hi Stephen
>  
> For bulk Cu you will need to apply
> -          k point sampling (e.g. MP 8 8 8)
> -          diagonalization (with Broyden mixing)
> -          smearing (e.g. T(elec) = 2000 K)
>  
> Moreover, you should use the conventional (cubic) unit cell with 4 Cu atoms instead of the primitive one with one Cu atom. CP2K is slower for non-orthorhombic cells.
> Plane wave (PW) codes are fast for small unit cells as they scale with the size of the cell which has to be filled with PWs. That’s not the case for CP2K using atomic basis functions which requires to calculate the interactions with the image Cu atoms in neighboring cells. So it can make sense to use multiple unit cells, e.g. a 2x2x2 supercell, and to reduce the k point mesh correspondingly (e.g. from 8x8x8 to 4x4x4).
>  
> Below you will find a CP2K input for bulk Cu in which I have considered the suggestions above. No guarantee, of course, that it will work properly.
>  
> Matthias
>  
> &GLOBAL
> PRINT_LEVEL low
> PROJECT_NAME Cu
> RUN_TYPE cell_opt
> &END GLOBAL
>  
> &MOTION
> &CELL_OPT
>   EXTERNAL_PRESSURE [bar] 1.0
>   MAX_DR 0.001
>   MAX_FORCE 0.0001
>   MAX_ITER 400
>   OPTIMIZER BFGS
>   PRESSURE_TOLERANCE [bar] 10.0
>   RMS_DR 0.0003
>   RMS_FORCE 0.00003
>   TYPE direct_cell_opt
>   &BFGS
>    TRUST_RADIUS 0.1
>    USE_MODEL_HESSIAN off
>    USE_RAT_FUN_OPT on
>   &END BFGS
> &END CELL_OPT
> &END MOTION
>  
> &FORCE_EVAL
> METHOD QS
> STRESS_TENSOR analytical
> &DFT
>   BASIS_SET_FILE_NAME BASIS_MOLOPT
>   POTENTIAL_FILE_NAME GTH_POTENTIALS
>   &KPOINTS
>    SCHEME MONKHORST-PACK 2 2 2
>    FULL_GRID yes
>    SYMMETRY yes
>    VERBOSE yes
>    PARALLEL_GROUP_SIZE -1
>   &END KPOINTS
>   &MGRID
>    NGRIDS 5
>    CUTOFF 400.0
>    REL_CUTOFF 60.0
>   &END MGRID
>   &QS
>    EPS_DEFAULT 1.0E-12
>    EXTRAPOLATION use_prev_p
>   &END QS
>   &SCF
>    ADDED_MOS 60
>    EPS_SCF 1.0E-8
>    MAX_SCF 300
>    SCF_GUESS restart
>    &DIAGONALIZATION yes
>     ALGORITHM STANDARD
>    &END DIAGONALIZATION
>    &MIXING yes
>     ALPHA 0.4
>     BETA 1.0
>     METHOD broyden_mixing
>     NBROYDEN 8
>    &END MIXING
>    &SMEAR on
>     METHOD FERMI_DIRAC
>     ELECTRONIC_TEMPERATURE [K] 2000.0
>    &END SMEAR
>   &END SCF
>   &XC
>    &XC_FUNCTIONAL PBE
>    &END XC_FUNCTIONAL
>    &VDW_POTENTIAL
>     POTENTIAL_TYPE pair_potential
>     &PAIR_POTENTIAL
>      TYPE DFTD3(BJ)
>      PARAMETER_FILE_NAME dftd3.dat
>      REFERENCE_FUNCTIONAL PBE
>     &END PAIR_POTENTIAL
>    &END VDW_POTENTIAL
>   &END XC
> &END DFT
> &SUBSYS
>   &CELL
>    ABC 3.62 3.62 3.62
>    MULTIPLE_UNIT_CELL 2 2 2
>   &END CELL
>   &COORD
>    SCALED
>    Cu    0    0    0
>    Cu    0  1/2  1/2
>    Cu  1/2    0  1/2
>    Cu  1/2  1/2    0
>   &END COORD
>   &KIND Cu
>    BASIS_SET DZVP-MOLOPT-SR-GTH-q11
>    POTENTIAL GTH-PBE-q11
>   &END KIND
>   &TOPOLOGY
>    MULTIPLE_UNIT_CELL 2 2 2
>   &END TOPOLOGY
> &END SUBSYS
> &END FORCE_EVAL
>  
> From: cp... at googlegroups.com <cp... at googlegroups.com> On Behalf Of Stephen Vicchio
> Sent: Dienstag, 15. September 2020 17:06
> To: cp... at googlegroups.com
> Subject: Re: [CP2K:13894] Need advice on bulk metal system calculations
>  
> From materials project, the .cif file I'm using for bulk Cu has unit cell dimensions of 2.24 2.24 2.24 and 60 60 60. Do you think that might explain what's happening? The error message I'm receiving isn't located in the out file from CP2K, but the out file from the cluster:
>  
> Error termination. Backtrace:
> At line 470 of file /tmp/spack/spack-stage/spack-stage-cp2k-7.1-h7ky2rvmv3fvzjrtpbqv2rr3fie3l6zy/spack-src/src/cell_methods.F
> Fortran runtime error: Bad real number in item 1 of list input
>  
> The example scripts run fines, but it my current .inp that is giving me issues. 
> 
> Stephen 
>  
> On Tue, Sep 15, 2020 at 11:01 AM Marcella Iannuzzi <marci... at gmail.com> wrote:
> Dear Stephen
>  
> Are you sure about 
>  ALPHA_BETA_GAMMA 60 60 60 
>  
> Best regards
> Marcella
> 
> On Tuesday, September 15, 2020 at 4:22:21 PM UTC+2 Stephen wrote:
> Hi Matthias, 
>  
> Thanks for your input! Using the CP2K example 'Getting the band structure of graphene' (https://www.cp2k.org/exercises:2016_uzh_cmest:band_structure_calculation), I was able to add the KPOINT section into my current input shown above. I've been altering the parameters and playing around with the input file, but I'm having issues troubleshooting why my new input isn't working properly.  The CP2K example was able to run, but my job isn't which makes me think there is an issue with some of the other flags I've set in my .inp file for bulk Cu. 
>  
> Below you can find the new .inp file that I've been using. Thanks for taking a look! 
>  
> With much appreciation, 
> 
> Stephen 
> 
> &GLOBAL
>    PRINT_LEVEL  MEDIUM
>    PROJECT_NAME bulk-Cu-CELL_DIMA
>    RUN_TYPE  ENERGY_FORCE 
>    WALLTIME  23:40:00 
> &END GLOBAL
>  &FORCE_EVAL
>    METHOD  QS
>    STRESS_TENSOR  ANALYTICAL
> 
>    &DFT
>      BASIS_SET_FILE_NAME BASIS_file
>      POTENTIAL_FILE_NAME POTENTIALS_file
>      UKS  T 
>      MULTIPLICITY  2 
>      CHARGE  0
>      &SCF
>        MAX_SCF  1000
>        EPS_SCF     9.9999999999999995E-07
>        SCF_GUESS  ATOMIC
>        &OT  T
>          MINIMIZER  CG
>          PRECONDITIONER  FULL_ALL
>          ENERGY_GAP     1.0000000000000000E-03
>        &END OT
>        &OUTER_SCF  T
>          EPS_SCF     9.9999999999999995E-07
>          MAX_SCF  50
>        &END OUTER_SCF
>      &END SCF
>      &QS
>        EPS_DEFAULT     1.0000000000000000E-10
>        METHOD  GPW
>        EXTRAPOLATION USE_GUESS
> 
>      &END QS
>      &MGRID
>        NGRIDS  5
>        CUTOFF     3.6000000000000000E+02
>        REL_CUTOFF     8.0000000000000000E+01
>      &END MGRID
>      &XC
>        DENSITY_CUTOFF     1.0000000000000000E-10
>        GRADIENT_CUTOFF     1.0000000000000000E-10
>        TAU_CUTOFF     1.0000000000000000E-10
>        &XC_FUNCTIONAL  NO_SHORTCUT
>          &PBE  T
>          &END PBE
>        &END XC_FUNCTIONAL
>        &VDW_POTENTIAL
>          POTENTIAL_TYPE  PAIR_POTENTIAL
>          &PAIR_POTENTIAL
>            TYPE  DFTD3(BJ)
>            PARAMETER_FILE_NAME dftd3.dat
>            REFERENCE_FUNCTIONAL PBE
>            CALCULATE_C9_TERM  F
>          &END PAIR_POTENTIAL
>        &END VDW_POTENTIAL
>      &END XC
>      &KPOINTS
>        SCHEME MONKHORST-PACK 1 1 1
>        SYMMETRY OFF 
>        WAVEFUNCTIONS REAL
>        FULL_GRID .TRUE.
>        PARALLEL_GROUP_SIZE  0     
>      &END KPOINTS
>    &END DFT
>    &SUBSYS
>      &CELL
>        ABC  2.24 2.24 2.24 
>        ALPHA_BETA_GAMMA 60 60 60 
>        MULTIPLE_UNIT_CELL  1 1 1
>      &END CELL
>      &TOPOLOGY
>        COORD_FILE_FORMAT CIF
>        COORD_FILE_NAME bulk-Cu.cif
>        NUMBER_OF_ATOMS  1
>        MULTIPLE_UNIT_CELL  1 1 1
>      &END TOPOLOGY
>      &KIND Cu 
>        BASIS_SET DZVP-MOLOPT-SR-GTH-q11
>        POTENTIAL GTH-PBE-q11
>      &END KIND
>    &END SUBSYS
>    &PRINT
>      &FORCES  ON
>      &END FORCES
>    &END PRINT
>  &END FORCE_EVAL
>  
>  
> On Mon, Sep 14, 2020 at 11:27 AM Krack Matthias (PSI) <mat... at psi.ch> wrote:
> Hi Stephen
>  
> For the description of metallic bulk systems, you will need Brillouin zone sampling with an appropriate k point mesh. It seems this is missing in your input.
>  
> HTH
>  
> Matthias
>  
> From: c... at googlegroups.com <c... at googlegroups.com> On Behalf Of Stephen
> Sent: Montag, 14. September 2020 14:54
> To: cp2k <c... at googlegroups.com>
> Subject: [CP2K:13869] Need advice on bulk metal system calculations
>  
> Hi All, 
>  
> I was wondering if anyone has tips for computing bulk metallic system in CP2K (specifically for bulk Cu and bulk Ni system). I need the electronic energy for bulk Cu and bulk Ni for some modeling on NP thermodynamic stability. 
>  
> I've optimized the unit cell parameters for both Cu and Ni systems by varying the unit cell dimensions found on Materials Project with the same level of theory as my CP2K on metal-organic frameworks. When I try to compute different metallic properties though such as atomization and bulk Cu + O2 (g) --> bulk CuO my energies are way off.. 
> From Web Elements for Ni, the atomization energy was determined:
> 
> bulk Cu (s) --> Cu(0) (g) w/ deltaH=338 kJ/mol
> 
> Now using DFT, the follow atomisation energy was determined to be:
> 
> bulk Cu (s) --> Cu(0) (g) w/ deltaE=1658.2 kJ/mol
>  
> Either my bulk system is too stable, or my atomized Cu is too unstable. Here is my DFT section for the bulk Cu calculations. 
>  
>    &DFT
>      BASIS_SET_FILE_NAME BASIS_file
>      POTENTIAL_FILE_NAME POTENTIALS_file
>      UKS  T 
>      MULTIPLICITY  2 
>      CHARGE  0
>      &SCF
>        MAX_SCF  1000
>        EPS_SCF     9.9999999999999995E-07
>        SCF_GUESS  ATOMIC
>        &OT  T
>          MINIMIZER  CG
>          PRECONDITIONER  FULL_ALL
>          ENERGY_GAP     1.0000000000000000E-03
>        &END OT
>        &OUTER_SCF  T
>          EPS_SCF     9.9999999999999995E-07
>          MAX_SCF  50
>        &END OUTER_SCF
>      &END SCF
>      &QS
>        EPS_DEFAULT     1.0000000000000000E-10
>        METHOD  GPW
>      &END QS
>      &MGRID
>        NGRIDS  5
>        CUTOFF     3.6000000000000000E+02
>        REL_CUTOFF     8.0000000000000000E+01
>      &END MGRID
>      &XC
>        DENSITY_CUTOFF     1.0000000000000000E-10
>        GRADIENT_CUTOFF     1.0000000000000000E-10
>        TAU_CUTOFF     1.0000000000000000E-10
>        &XC_FUNCTIONAL  NO_SHORTCUT
>          &PBE  T
>          &END PBE
>        &END XC_FUNCTIONAL
>        &VDW_POTENTIAL
>          POTENTIAL_TYPE  PAIR_POTENTIAL
>          &PAIR_POTENTIAL
>            TYPE  DFTD3(BJ)
>            PARAMETER_FILE_NAME dftd3.dat
>            REFERENCE_FUNCTIONAL PBE
>            CALCULATE_C9_TERM  F
>          &END PAIR_POTENTIAL
>        &END VDW_POTENTIAL
>      &END XC
>    &END DFT
>  
> Sample .inp and .out files are attached. 
>  
> Any feedback would be much appreciated! Thanks for taking a look at my issues. 
>  
> With much appreciation, 
> 
> Stephen 
> -- 
> 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 cp... at googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/a3f838aa-94cb-4998-b501-09656b8efd2en%40googlegroups.com.
> --
> You received this message because you are subscribed to a topic in the Google Groups "cp2k" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/topic/cp2k/g7wXQbqxfQs/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to cp... at googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/734dfa8497b94d2fb178deb4655bfee5%40psi.ch.
> 
>  
> --
> Stephen P. Vicchio
> PhD Candidate | Getman Research Group
> Dept. of Education GAANN Fellow
> Department of Chemical & Biomolecular Engineering
> 105 Earle Hall | Clemson, SC 29634-0909
> s... at g.clemson.edu
> 
> -- 
> You received this message because you are subscribed to a topic in the Google Groups "cp2k" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/topic/cp2k/g7wXQbqxfQs/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to cp... at googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/b3ea4a80-54af-4ac4-b5e7-a4922f4762dan%40googlegroups.com.
> 
>  
> --
> Stephen P. Vicchio
> PhD Candidate | Getman Research Group
> Dept. of Education GAANN Fellow
> Department of Chemical & Biomolecular Engineering
> 105 Earle Hall | Clemson, SC 29634-0909
> svi... at g.clemson.edu
> -- 
> 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 cp... at googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CAHf%3DnNaw5e%2BioxzubJD7RNmia8OjAoekUOZYx_1bi-kvh4tSuQ%40mail.gmail.com.
> -- 
> 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 cp... at googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/75e08d6442354e1d9a1393a676b0aac2%40psi.ch.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200915/265dfddc/attachment.htm>


More information about the CP2K-user mailing list