Bulk modulus of silicon with CP2K

[leila]

Hi all, 

I simulated the silicon with 64 atoms and calculated the energy-volume plot and fitted with murnaghan.

I found the lattice parameter 5.48 A ( lattice parameter in experiment is 5.44 A), That is OK but the bulk modulus is 63 GPa that is very low. (B_exp = 100 GPa)

I really don' t know why cp2k gets B so low???

I tried with DZVP basis set and PBE functional with 280 Ryd cut off for density.

I also checked with TZVP, it is not changing so much. It is 66 GPa.

Can anybody help me with that?

You can find my inputfile in below.

&GLOBAL

  PROJECT si64-10.88A

  PRINT_LEVEL LOW

  RUN_TYPE GEO_OPT

&END GLOBAL

&MOTION

  &CELL_OPT

    TYPE GEO_OPT

    OPTIMIZER CG

    MAX_ITER 20

    EXTERNAL_PRESSURE [bar] 1.0

    MAX_DR 0.02

    RMS_DR 0.01

    MAX_FORCE 0.0002

    RMS_FORCE 0.0001

    KEEP_ANGLES

    &CG

      &LINE_SEARCH

        TYPE 2PNT

        &2PNT

        &END

      &END

    &END

  &END

  &GEO_OPT

    MAX_ITER 200

    MINIMIZER BFGS

  &END

  &PRINT

    &STRESS

    &END STRESS

    &RESTART

      &EACH

        CELL_OPT 1

      &END

    &END

  &END

&END

&FORCE_EVAL

  METHOD QS

  STRESS_TENSOR ANALYTICAL

  &DFT

    BASIS_SET_FILE_NAME ./BASIS_MOLOPT

    POTENTIAL_FILE_NAME ./GTH_POTENTIALS

    &MGRID

      CUTOFF 280

    &END MGRID

    &QS

      METHOD GPW

    &END QS

    &SCF

      SCF_GUESS RESTART

      MAX_SCF 20

      EPS_SCF 1.0e-5

      &OUTER_SCF

      EPS_SCF 1.0E-5

      &END

      &OT ON

#        MINIMIZER DIIS

         ROTATION

      &END OT

      &PRINT

        &RESTART

          BACKUP_COPIES 0

        &END RESTART

      &END PRINT

    &END SCF

    &XC

      &XC_FUNCTIONAL PBE

      &END XC_FUNCTIONAL

      &vdW_POTENTIAL

         DISPERSION_FUNCTIONAL PAIR_POTENTIAL

         &PAIR_POTENTIAL

            TYPE DFTD3

            CALCULATE_C9_TERM .TRUE.

            REFERENCE_C9_TERM .TRUE.

            LONG_RANGE_CORRECTION .TRUE.

            PARAMETER_FILE_NAME ./dftd3.dat

            REFERENCE_FUNCTIONAL PBE

            R_CUTOFF 8.

            EPS_CN 0.01

         &END PAIR_POTENTIAL

      &END vdW_POTENTIAL

    &END XC

  &END DFT

  &SUBSYS

    &TOPOLOGY

      COORD_FILE_NAME ./GEOMETRY.xyz

      COORDINATE XYZ

#      CONNECTIVITY OFF

    &END TOPOLOGY

    &KIND Si

      BASIS_SET DZVP-MOLOPT-SR-GTH

#      BASIS_SET SZV-MOLOPT-SR-GTH

#       BASIS_SET TZVP-MOLOPT-GTH

      POTENTIAL GTH-PBE-q4

    &END

    &CELL

      ABC    10.88  10.88 10.88

    &END

  &END SUBSYS

 &END

[hut...@pci.uzh.ch]

Hi

two things to do:
1) What is the official PBE value for this quantity?
2) Use a REF_CELL and a high enough cutoff (has to be tested)
in order to get a smooth enough pressure curve

regards

Juerg

--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut...@pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com
From: leila
Sent by: cp...@googlegroups.com
Date: 08/12/2013 11:57AM
Subject: [CP2K:4549] Bulk modulus of silicon with CP2K

--
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...@googlegroups.com.
To post to this group, send email to cp...@googlegroups.com.
Visit this group at http://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[leila salimi]

Dear Juerg, 

Thanks for your answer.

1) the PBE value for B is 89 GPa.

2) I also tried with higher cutoff and REF_CUTOFF, but it didn't change.

&MGRID

      NGRIDS 8

      CUTOFF 500

      REL_CUTOFF  60

    &END MGRID

[Süle Péter]

Dear Leila,

Is that necessary to use dispersion correction for
bulk modulus calculation of Si ?
In your submitted input file vdw is included:

� � � &vdW_POTENTIAL

� � � � �DISPERSION_FUNCTIONAL PAIR_POTENTIAL
� � � � �&PAIR_POTENTIAL
� � � � � � TYPE DFTD3
� � � � � � CALCULATE_C9_TERM .TRUE.
� � � � � � REFERENCE_C9_TERM .TRUE.
� � � � � � LONG_RANGE_CORRECTION .TRUE.
� � � � � � PARAMETER_FILE_NAME ./dftd3.dat

I'm not sure, but it could influence the PBE results.

Best regard, P�ter S�le

> �
> �
>
>

[hut...@pci.uzh.ch]

Hi

I would guess that your calculation is not yet converged
with respect to

1) the basis set
2) the k-points. I would use a small unit cell and then
the MULTIPLE_UNIT_CELL keyword to check for convergence
3) the cutoff, you are probably close here (REL_CUTOFF 80, just to be sure)
I would still use the REF_CELL keyword in order to do constant
number of grid point calculations
4) the dispersion correction will change the reference PBE value

regards

Juerg


--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut...@pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com

From: leila salimi
Sent by: cp...@googlegroups.com
Date: 08/12/2013 03:05PM
Subject: Re: [CP2K:4550] Bulk modulus of silicon with CP2K

[leila salimi]

Hi Juerg, 

Thanks very much. I tried without dispersion correction for DZVP basis set, CUTOFF 500 and REL_CUTOFF 80. I got better results( B = 90 GPa).

I didn't understand how I can find my calculation is converged or not?

For k-points you mean the system of 64 atoms is not big enough and I have to check with small unit cell (8 atoms) and then with MULTIPLE_UNIT_CELL keyword increase the size of  the system and to see how it is converged?

 

Regards, 

Leila