Grimme Dispersion with SCC DFTB

[Jan-Niclas Luy]

Hello,

I have a question regarding dispersion corrections for the SCC DFTB method implemented in CP2K.

I tell the program to use Dispersion correction, but the line in the output shows only zeroes like:

  Core Hamiltonian energy:                                     -4.17194433927250
  Repulsive potential energy:                                   0.07921577780740
  Electronic energy:                                            0.02107155371499
  Dispersion energy:                                            0.00000000000000

  Total energy:                                                -4.07165700775011

 ENERGY| Total FORCE_EVAL ( QS ) energy (a.u.):               -4.071657007750285

My input for a simple water molecule looks like:

&GLOBAL
  PROJECT H2O
  RUN_TYPE GEO_OPT
  PRINT_LEVEL LOW
&END GLOBAL

&FORCE_EVAL
  METHOD QS
  &DFT
    &MGRID
      CUTOFF 500
    &END MGRID
    &QS
      METHOD DFTB
      &DFTB
        SELF_CONSISTENT    T
        DO_EWALD           T
        DISPERSION        
        &PARAMETER
          PARAM_FILE_PATH  /home/Luyj/test/cp2k510/data/DFTB/scc/
          PARAM_FILE_NAME  scc_parameter
          DISPERSION_PARAMETER_FILE /home/Luyj/test/cp2k510/data/dftd3.dat
          DISPERSION_TYPE D3
          #UFF_FORCE_FIELD  uff_table
        &END PARAMETER
      &END DFTB
    &END QS
    &SCF
      MAX_SCF 30
      EPS_SCF 1.0E-6
      SCF_GUESS ATOMIC
      &OT
        MINIMIZER DIIS
        PRECONDITIONER FULL_SINGLE_INVERSE
      &END
    &END SCF
  &END DFT
  &SUBSYS
    &CELL
      ABC 12.4138 12.4138 12.4138
    &END CELL
    &COORD
      O      12.235322       1.376642      10.869880
      H      12.415139       2.233125      11.257611
      H      11.922476       1.573799       9.986994
    &END COORD
  &END SUBSYS
&END FORCE_EVAL

&MOTION
  &GEO_OPT
    TYPE MINIMIZATION
    MAX_DR    1.0E-03
    MAX_FORCE 1.0E-03
    RMS_DR    1.0E-03   
    RMS_FORCE 1.0E-03
    MAX_ITER 200
    OPTIMIZER CG
    &CG
      MAX_STEEP_STEPS  0
      RESTART_LIMIT 9.0E-01
    &END CG
  &END GEO_OPT
  &CONSTRAINT
    &FIXED_ATOMS
      COMPONENTS_TO_FIX XYZ
      LIST 1
    &END FIXED_ATOMS
  &END CONSTRAINT
&END MOTION


Is something wrong with my input or is parameter set not supported?

[hut...@chem.uzh.ch]

Hi

you have to set the D3_SCALING parameters (default is 0.0, see manual).

regards

Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: cp2k <cp...@googlegroups.com>
From: Jan-Niclas Luy
Sent by: cp...@googlegroups.com
Date: 05/07/2018 01:07PM
Subject: [CP2K:10286] Grimme Dispersion with SCC DFTB

--
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 https://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/d/optout.

[Jan-Niclas Luy]

Thank you for the reply.

Unfortunately the problem persists.

I added the following line to above input, but the dispersion energy is still zero.

          D3_SCALING  1 1 1

Best regards
Jan-Niclas

[hut...@chem.uzh.ch]

Hi

does the test (in the regtest directory)

tests/DFTB/regtest-scc/h2o_disp3.inp

work?

Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: cp2k <cp...@googlegroups.com>
From: Jan-Niclas Luy
Sent by: cp...@googlegroups.com

Date: 05/08/2018 10:24AM
Subject: Re: [CP2K:10289] Grimme Dispersion with SCC DFTB

[Jan-Niclas Luy]

Hi,

the test calculation worked. By copying the DFTB section I got my example to work as well. I am not quite sure what caused the error. Could it have been an old .restart file still present in the directory?

[njust...@gmail.com]

hello，

Do you have any idea on how to set the D3_SCALING parameters?  I use CP2K to optimize the crystal structure, but I didn't find the reference on how to set the Grimme D3 correction.

在 2018年5月7日星期一 UTC+8下午7:07:00，Jan-Niclas Luy写道：

[hut...@chem.uzh.ch]

Hi

the keyword

D3_SCALING s6 sr6 s8

in the parameter section is required. There are no defaults for these
parameters, or better defaults are all 0.
See also the manual.

regards

Juerg

--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: "cp2k" <cp...@googlegroups.com>

From: njust...@gmail.com
Sent by: cp...@googlegroups.com
Date: 01/21/2019 06:46AM
Subject: [CP2K:11203] Re: Grimme Dispersion with SCC DFTB