[CP2K-user] [CP2K:16137] Cell Optimization Not Converged for Copper

[Fabian Ducry]

Hi Geng,

The plane wave basis depends on the size of the unit cell. Therefore, a 
change of the cell dimensions corresponds to a change of basis. If the 
energy difference that comes from this change of basis is large the 
optimizer cannot converge. With CELL_REF you prevent this issue by 
fixing the basis (which is now independent of the volume of the 
simulation domain).

By the way, you are confusing the cell optimization and SCF: The SCF is 
the energy minimization at each step of the cell optimization (and this 
always converged). Only the cell optimization did not converge.

Regarding the EPS_SCF: a tighter value can lower the number of cell 
optimization steps, but each step will be more expensive. You have to 
test it yourself.

Cheers,

Fabian

On 27.10.2021 20:50, GENG YUAN wrote:
> Dear Fabian,
>
> Many thanks for your reply. I added CELL_REF in my simulation and it 
> works!
>
> Is it because that CELL_REF set a larger cell dimension, so the change 
> of my real cell volume during the cell_opt is always under this 
> reference cell volume, therefore the SCF could converge rather than 
> keep fluctuating? (please correct me if I'm wrong). I found some 
> information saying that CELL_REF is to fix the plane-wave basis whilst 
> doing a cell_opt, but I don't know what does this means (I will try 
> searching more).
>
> About the EPS_SCF, do you have any suggestions about what criteria 
> could I apply instead of the one I used in my simulation (1.0E-06)?
>
> Many thanks for your help!
> Sincerely,
> Geng.
>
> 在2021年10月27日星期三 UTC-4 上午3:42:53<fabia... at gmail.com> 写道：
>
>     Hi Geng,
>
>     Add a &CELL_REF section
>     (https://manual.cp2k.org/cp2k-8_2-branch/CP2K_INPUT/FORCE_EVAL/SUBSYS/CELL/CELL_REF.html).
>     With
>
>     &CELL_REF
>       ABC 10.83*1.5 10.83*1.5 7.22*1.5
>     &END
>
>     the optimization converges within 35 steps. You could decrease
>     EPS_SCF to reduce the noise in the forces.
>
>     Cheers,
>
>     Fabian
>
>     On 26.10.2021 23:02, GENG YUAN wrote:
>>     Dear CP2K Users,
>>
>>     I am attempting to run the cell optimization for a single bulk Cu
>>     (with 72 atoms). However, the SCF calculation can not converge:
>>     It reached the maximum number of optimization steps (500) and
>>     therefore exit the optimization.
>>
>>     I also tried changing the original lattice in my simulation (with
>>     associated coordinate) but it didn't work either.
>>
>>     So far, the cell optimization only converged when I use a smaller
>>     bulk Cu (with 32 atoms). But we'd like to run a larger system.
>>
>>     I've attached my input, output, the original Cu coordinate, as
>>     well as the trajectory during cell optimization here for your
>>     convenience. Any ideas are welcome!
>>
>>     Below is the input:
>>     -----------------------------
>>     &GLOBAL
>>       PROJECT Cu_cellopt
>>       RUN_TYPE CELL_OPT
>>       PRINT_LEVEL LOW
>>     &END GLOBAL
>>     &FORCE_EVAL
>>       METHOD QS
>>       STRESS_TENSOR ANALYTICAL
>>       &SUBSYS
>>         &CELL
>>           ABC 10.83 10.83 7.22
>>           SYMMETRY TETRAGONAL
>>         &END CELL
>>         &TOPOLOGY
>>          COORD_FILE_NAME ./Cu332.xyz
>>          COORDINATE xyz
>>         &END
>>         &KIND Cu
>>           BASIS_SET DZVP-MOLOPT-SR-GTH
>>           POTENTIAL GTH-PBE-q11
>>         &END KIND
>>       &END SUBSYS
>>       &DFT
>>         BASIS_SET_FILE_NAME ./BASIS_MOLOPT
>>         POTENTIAL_FILE_NAME ./GTH_POTENTIALS
>>         &QS
>>           EPS_DEFAULT 1.0E-12
>>         &END QS
>>         &MGRID
>>           CUTOFF 400
>>           NGRIDS 5
>>           REL_CUTOFF 60
>>         &END MGRID
>>         &SCF
>>           SCF_GUESS ATOMIC
>>           EPS_SCF 1.0E-06
>>           MAX_SCF 200
>>           ADDED_MOS 100
>>           &OUTER_SCF
>>             MAX_SCF 50
>>             EPS_SCF 1.0E-6
>>           &END
>>           &DIAGONALIZATION T
>>             ALGORITHM STANDARD
>>           &END DIAGONALIZATION
>>           &MIXING T
>>             METHOD BROYDEN_MIXING
>>             ALPHA 0.1
>>             NBROYDEN 8
>>           &END MIXING
>>           &SMEAR ON
>>             METHOD FERMI_DIRAC
>>             ELECTRONIC_TEMPERATURE [K] 500
>>           &END SMEAR
>>           &PRINT
>>             &RESTART ON
>>             &END RESTART
>>           &END PRINT
>>         &END SCF
>>         &XC
>>           &XC_FUNCTIONAL
>>              &LIBXC
>>                FUNCTIONAL XC_GGA_X_PBE_R
>>              &END
>>              &LIBXC
>>                FUNCTIONAL XC_GGA_C_PBE
>>              &END
>>           &END XC_FUNCTIONAL
>>           &VDW_POTENTIAL
>>              POTENTIAL_TYPE PAIR_POTENTIAL
>>              &PAIR_POTENTIAL
>>                 LONG_RANGE_CORRECTION
>>                 PARAMETER_FILE_NAME dftd3.dat
>>                 TYPE DFTD3
>>                 REFERENCE_FUNCTIONAL revPBE
>>              &END PAIR_POTENTIAL
>>            &END VDW_POTENTIAL
>>            &XC_GRID
>>             XC_DERIV NN50_SMOOTH
>>            &END XC_GRID
>>         &END XC
>>       &END DFT
>>     &END FORCE_EVAL
>>     &MOTION
>>       &CELL_OPT
>>         EXTERNAL_PRESSURE 0
>>         TYPE DIRECT_CELL_OPT
>>         KEEP_SYMMETRY TRUE
>>         MAX_DR    1.0E-05
>>         MAX_FORCE 1.0E-05
>>         RMS_DR    1.0E-05
>>         RMS_FORCE 1.0E-05
>>         MAX_ITER 500
>>         OPTIMIZER BFGS
>>       &END CELL_OPT
>>     &END MOTION
>>     ------------------------------------
>>     Many thanks,
>>     Geng.
>>
>>     -- 
>>     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+uns... at googlegroups.com.
>>     To view this discussion on the web visit
>>     https://groups.google.com/d/msgid/cp2k/6e75296d-eecd-4691-8e6b-560464b59bebn%40googlegroups.com
>>     <https://groups.google.com/d/msgid/cp2k/6e75296d-eecd-4691-8e6b-560464b59bebn%40googlegroups.com?utm_medium=email&utm_source=footer>.
>
> -- 
> 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/d4b2ad27-1cea-4e3c-97eb-f49f52fe0ebcn%40googlegroups.com 
> <https://groups.google.com/d/msgid/cp2k/d4b2ad27-1cea-4e3c-97eb-f49f52fe0ebcn%40googlegroups.com?utm_medium=email&utm_source=footer>.

-- 
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/2c9545a3-9f90-5058-348d-f8c0d5072393%40gmail.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20211028/f943acd7/attachment.htm>