TDDFT second derivatives

119 views
Skip to first unread message

andres.ort...@epfl.ch

unread,
Sep 16, 2020, 6:26:16 AM9/16/20
to cp2k
Dear CP2K Users, 
I was wondering if you could help me with something
i am using the CP2K 7.1 version , I am trying to do TDDFT calculations. 
however I get this error 

 *******************************************************************************

 *   ___                                                                       *

 *  /   \                                                                      *

 * [ABORT]                                                                     *

 *  \___/                derivatives bigger than 2 not implemented             *

 *    |                                                                        *

 *  O/|                                                                        *

 * /| |                                                                        *

 * / \                                            xc/xc_xpbe_hole_t_c_lr.F:198 *

 *******************************************************************************



However in my input i am not considering the PBE HOLE functional and I have included the derivatives.  I.was wondering if this is an issue of the version 7?


best.


Andres 

     &XC

       DENSITY_CUTOFF     1.0000000000000000E-10

       GRADIENT_CUTOFF     1.0000000000000000E-10

       TAU_CUTOFF     1.0000000000000000E-10

       &XC_FUNCTIONAL

         &PBE  T

           SCALE_X     7.5000000000000000E-01

           SCALE_C     1.0000000000000000E+00

         &END PBE

       &END XC_FUNCTIONAL

       &HF

         FRACTION     2.5000000000000000E-01

         &SCREENING

           EPS_SCHWARZ     9.9999999999999995E-07

           SCREEN_ON_INITIAL_P  F

         &END SCREENING

         &INTERACTION_POTENTIAL

           POTENTIAL_TYPE  TRUNCATED

           CUTOFF_RADIUS     6.0000000000000000E+00

           T_C_G_DATA t_c_g.dat

         &END INTERACTION_POTENTIAL

         &MEMORY

           EPS_STORAGE_SCALING     1.0000000000000001E-01

           MAX_MEMORY  120000

         &END MEMORY

       &END HF

       &XC_GRID

        XC_DERIV SPLINE2_SMOOTH      # this is needed for the 2nd derivatives of the XC functional

      &END XC_GRID

       &VDW_POTENTIAL

         POTENTIAL_TYPE  PAIR_POTENTIAL

         &PAIR_POTENTIAL

           R_CUTOFF     1.0000000000000005E+01

           TYPE  DFTD3(BJ)

           PARAMETER_FILE_NAME dftd3.dat

           REFERENCE_FUNCTIONAL PBE0

           D3BJ_SCALING 1.000   0.4145  1.2177  4.8593

         &END PAIR_POTENTIAL

       &END VDW_POTENTIAL

     &END XC


hut...@chem.uzh.ch

unread,
Sep 16, 2020, 9:18:46 AM9/16/20
to cp...@googlegroups.com
Hi
your &XC section works for me using the current trunk version.
I don't know if it's another part of your input or a recent fix
in the code that is responsible.

regards

Juerg Hutter
--------------------------------------------------------------
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: "andres.ort...@epfl.ch"
Sent by: cp...@googlegroups.com
Date: 09/16/2020 12:26PM
Subject: [CP2K:13908] TDDFT second derivatives
--
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 view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/f4452568-e0c0-436e-b4e1-9511e61b63b7n%40googlegroups.com.

HB H

unread,
Sep 19, 2020, 8:39:00 PM9/19/20
to cp2k
Hi man,

I got the same error while running with B3LYP (with the similar &XC section), looks like the &HF section does not support sec-derivatives ? 

Let me know if you have solved this issue or not.

Thx
Haibei

Andres Adolfo Ortega Guerrero

unread,
Sep 20, 2020, 3:44:17 AM9/20/20
to cp2k
Dear Prof, Hutter. 

thank you for your email, I tried my input in the cp2k 5.1 and 6.1 and it worked fine. I guess it should be an issue in the CP2K 7.1 version in daint. 


best, 

Andres 

Matt W

unread,
Sep 21, 2020, 7:32:21 AM9/21/20
to cp2k
As a workaround for now if you add

 EXCH_CORRECTION_FUNC PBEX

into the ADMM section the calculation runs (at least in the developer version).

Matt

Reply all
Reply to author
Forward
0 new messages