[CP2K-user] [CP2K:18110] Re: Not getting correct optimised geometry for Au(111) surface with ligands (BDT or DMBT)
Vikash Khokhar
vikashkhokhar141 at gmail.com
Tue Nov 29 12:22:42 UTC 2022
Thank you, Marcella
I'll look into this.
Regards
Vikash
On Tuesday, November 29, 2022 at 5:45:58 PM UTC+5:30 Marcella Iannuzzi
wrote:
> Dear Vikash
>
> In the posted input, the k-points are not used.
> If you already used them, try with a denser mesh and with the *MONKHORST-PACK
> scheme.*
> I suggest that you test on one configuration (possibly already a bit
> distorted) the effects on energies and forces by changing k-point mesh and
> cutoff.
> If energies and forces are not accurate, the geometry will be wrong.
> EPS_SCF 1.0E-07 should be good enough.
> OUTER_SCF is used only to reinitialise the preconditioner for the OT
> approach, hence it is useless with the standard diagonalization method.
> 3 layers for a Au slab are insufficient and are keen to lead to weird
> distortions in the lattice.
>
> Regards
> Marcella
>
>
>
> On Tuesday, November 29, 2022 at 5:58:54 AM UTC+1 vikashkh... at gmail.com
> wrote:
>
>> Thank you, Marcella,
>> I would like to know the exact PW cutoff I can use.
>> And yes, I have used mesh k-points 2 3 1, are these correct, or do I
>> need to take a bigger grid? If yes, what should it be?
>> I thought the default tolerance would be okay. If not, then how much
>> tolerance would be enough?
>>
>> Regards
>> Vikash
>> On Monday, November 28, 2022 at 8:17:19 PM UTC+5:30 Marcella Iannuzzi
>> wrote:
>>
>>> dear Vikash
>>>
>>> The system is probably too small.
>>> To stabilise the Au slab one should use more layers, we use 6.
>>> With such small lateral dimension you should use a mesh of k-points.
>>> Moreover, the tolerance for the convergence of the wavefunction is not
>>> very tight and the PW cutoff seems quite low.
>>>
>>> Regards
>>> Marcella
>>>
>>> On Monday, November 28, 2022 at 1:58:13 PM UTC+1 vikashkh... at gmail.com
>>> wrote:
>>>
>>>> I'm trying to optimize a small unit cell of Au(111) surface with
>>>> ligands (BDT or DMBT) but I'm not getting the correct optimized geometry.
>>>> I'm using Cp2k 8.1 version.
>>>> I'm attaching the Input file here, please have a look and suggest to me
>>>> what's wrong with my calculation.
>>>>
>>>> &GLOBAL
>>>> PROJECT Au_64_2BDT
>>>> RUN_TYPE GEO_OPT
>>>> PRINT_LEVEL MEDIUM
>>>> &END GLOBAL
>>>>
>>>> &MOTION
>>>> &GEO_OPT
>>>> TYPE MINIMIZATION
>>>> OPTIMIZER BFGS
>>>> MAX_DR 3.00E-03
>>>> MAX_FORCE 4.50E-04
>>>> RMS_DR 1.50E-03
>>>> RMS_FORCE 3.00E-04
>>>> MAX_ITER 3000
>>>> &END GEO_OPT
>>>>
>>>> &CONSTRAINT
>>>> &FIXED_ATOMS
>>>> COMPONENTS_TO_FIX XYZ
>>>> LIST 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
>>>> 24
>>>> &END FIXED_ATOMS
>>>> &END CONSTRAINT
>>>>
>>>> &END MOTION
>>>>
>>>> &FORCE_EVAL
>>>> METHOD QS
>>>>
>>>> &DFT
>>>> BASIS_SET_FILE_NAME BASIS_MOLOPT
>>>> POTENTIAL_FILE_NAME GTH_POTENTIALS
>>>>
>>>> &QS
>>>> METHOD GPW
>>>> EXTRAPOLATION USE_GUESS
>>>> EPS_DEFAULT 1.0E-10
>>>> &END QS
>>>>
>>>> &MGRID
>>>> CUTOFF 350
>>>> NGRIDS 4
>>>> REL_CUTOFF 60
>>>> &END MGRID
>>>>
>>>> &SCF
>>>> SCF_GUESS ATOMIC
>>>> EPS_SCF 1.0E-05
>>>> MAX_SCF 200
>>>> ADDED_MOS 10
>>>>
>>>> &OUTER_SCF
>>>> EPS_SCF 1.0E-04
>>>> MAX_SCF 200
>>>> &END OUTER_SCF
>>>>
>>>> &DIAGONALIZATION T
>>>> ALGORITHM STANDARD
>>>> &END DIAGONALIZATION
>>>>
>>>> &SMEAR ON
>>>> METHOD FERMI_DIRAC
>>>> ELECTRONIC_TEMPERATURE [K] 300
>>>> &END SMEAR
>>>>
>>>> &MIXING T
>>>> METHOD BROYDEN_MIXING
>>>> ALPHA 0.4
>>>> NBROYDEN 8
>>>> &END MIXING
>>>> &END SCF
>>>>
>>>> &XC
>>>> &XC_FUNCTIONAL PBE
>>>> &END XC_FUNCTIONAL
>>>>
>>>> &vdW_POTENTIAL
>>>> DISPERSION_FUNCTIONAL PAIR_POTENTIAL
>>>> &PAIR_POTENTIAL
>>>> PARAMETER_FILE_NAME dftd3.dat
>>>> TYPE DFTD3
>>>> REFERENCE_FUNCTIONAL PBE
>>>> R_CUTOFF 25.0
>>>> &END PAIR_POTENTIAL
>>>> &END vdW_POTENTIAL
>>>> &END XC
>>>>
>>>> &KPOINTS
>>>> SCHEME GAMMA 1 1 1
>>>> FULL_GRID .TRUE.
>>>> PARALLEL_GROUP_SIZE 0
>>>> &END KPOINTS
>>>>
>>>> &END DFT
>>>>
>>>> &SUBSYS
>>>> &CELL
>>>> PERIODIC XYZ
>>>> A 17.3099747000000015 0.0000000000000000
>>>> 0.0000000000000000
>>>> B 0.0000000000000000 9.9939184000000001
>>>> 0.0000000000000000
>>>> C 0.0000000000000000 0.0000000000000000
>>>> 30.0000000000000000
>>>> &END CELL
>>>>
>>>> &COORD
>>>> @INCLUDE Au_64_2BDT.xyz
>>>> &END COORD
>>>>
>>>> &KIND Au
>>>> BASIS_SET DZVP-MOLOPT-SR-GTH
>>>> POTENTIAL GTH-PBE
>>>> &END KIND
>>>> &KIND H
>>>> BASIS_SET DZVP-MOLOPT-GTH
>>>> POTENTIAL GTH-PBE
>>>> &END KIND
>>>> &KIND C
>>>> BASIS_SET DZVP-MOLOPT-GTH
>>>> POTENTIAL GTH-PBE
>>>> &END KIND
>>>> &KIND S
>>>> BASIS_SET DZVP-MOLOPT-GTH
>>>> POTENTIAL GTH-PBE
>>>> &END KIND
>>>> &END SUBSYS
>>>> &END FORCE_EVAL
>>>>
>>>> I'm also attaching the input and output geometry pics. In the input 3rd
>>>> layer of the gold surface is fixed and the above 2 layers optimize above
>>>> each other(which breaks 111 symmetry) which is not expected.
>>>> I hope you understand my problem and I expect a solution, earliest
>>>> possible.
>>>>
>>>> Regards
>>>> Vikash Khokhar
>>>> Indian Institute of Technology, Madras [image: Screenshot from
>>>> 2022-11-28 18-24-10.png][image: Screenshot from 2022-11-28
>>>> 18-25-47.png]
>>>>
>>>
--
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/956eff25-c8ec-41d3-b05b-ba73d1885208n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221129/1d4943c9/attachment-0001.htm>
More information about the CP2K-user
mailing list