TDDFT periodic with Hybrid functional

[oandr...@gmail.com]

Dear Cp2k devs and users, 

I am trying to do a tddft calculation for a periodic system  using a modified hybrid functional where HF is 15%

&AUXILIARY_DENSITY_MATRIX_METHOD
       ADMM_PURIFICATION_METHOD NONE
       METHOD BASIS_PROJECTION
       EXCH_SCALING_MODEL NONE
       EXCH_CORRECTION_FUNC PBEX
     &END AUXILIARY_DENSITY_MATRIX_METHOD
     &QS
       EPS_PGF_ORB  1.0000000000000001E-032
       METHOD GPW
     &END QS
     &MGRID
       NGRIDS 5
       CUTOFF  7.0000000000000000E+002
       REL_CUTOFF  7.0000000000000000E+001
     &END MGRID
     &XC
       DENSITY_CUTOFF  1.0000000000000000E-010
       GRADIENT_CUTOFF  1.0000000000000000E-010
       TAU_CUTOFF  1.0000000000000000E-010
       &XC_FUNCTIONAL NO_SHORTCUT
         &PBE T
           SCALE_X  8.4999999999999998E-001
           SCALE_C  1.0000000000000000E+000
         &END PBE
         !&PBE_HOLE_T_C_LR T
         !  SCALE_X  1.4999999999999999E-001
         !  CUTOFF_RADIUS  8.0000000000000000E+000
         !&END PBE_HOLE_T_C_LR
       &END XC_FUNCTIONAL
       &HF
         FRACTION  1.5000000000000022E-001
         &SCREENING
           EPS_SCHWARZ  9.9999999999999995E-008
           SCREEN_ON_INITIAL_P F
         &END SCREENING
         &INTERACTION_POTENTIAL
           POTENTIAL_TYPE TRUNCATED
           CUTOFF_RADIUS  8.0000000000000000E+000

       T_C_G_DATA "t_c_g.dat"
         &END INTERACTION_POTENTIAL
         &MEMORY
           EPS_STORAGE_SCALING  1.0000000000000001E-001
           MAX_MEMORY 40000
         &END MEMORY
       &END HF
     &END XC

I was wondering If I should define the XC_FUNCTIONAL in the TDDFPT section , or if like this is enough 

 &PROPERTIES
      &TDDFPT                        
       KERNEL FULL                       
      NSTATES 10                  
       MAX_ITER   100              
       CONVERGENCE [eV] 1.0e-7         
      ADMM_KERNEL_XC_CORRECTION T
       &MGRID                    
         CUTOFF 700                   
         REL_CUTOFF 70                
       &END MGRID
     &PRINT
       &NTO_ANALYSIS
        CUBE_FILES T
        STRIDE  3 3 3
       &END NTO_ANALYSIS
     &END PRINT
     &END TDDFPT
    &END PROPERTIES

or should i include 

     &XC
       &XC_FUNCTIONAL NO_SHORTCUT
         &PBE T
           SCALE_X  8.4999999999999998E-001
           SCALE_C  1.0000000000000000E+000
         &END PBE
         !&PBE_HOLE_T_C_LR T
         !  SCALE_X  1.4999999999999999E-001
         !  CUTOFF_RADIUS  8.0000000000000000E+000
         !&END PBE_HOLE_T_C_LR
       &END XC_FUNCTIONAL
       &HF
         FRACTION  1.5000000000000022E-001
         &SCREENING
           EPS_SCHWARZ  9.9999999999999995E-008
           SCREEN_ON_INITIAL_P F
         &END SCREENING
         &INTERACTION_POTENTIAL
           POTENTIAL_TYPE TRUNCATED
           CUTOFF_RADIUS  8.0000000000000000E+000
           T_C_G_DATA "t_c_g.dat"
         &END INTERACTION_POTENTIAL
       &END HF
     &END XC

In the XC definition of TDDFPT , 

Thank you for your suggestions and help

Andres Ortega-Guerrero

nanotech@surfaces | EMPA 

[Jürg Hutter]

Hi

you don't have to repeat the XC section within TDDFT if you want to
use the same functional as for the ground state.

The XC option in the TDDFT section allows to change the functional
for the kernel only. However, for hybrids/ADMM this is restricted.

regards
JH

________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of oandr...@gmail.com <oandr...@gmail.com>
Sent: Friday, March 7, 2025 7:43 AM
To: cp2k
Subject: [CP2K:21266] TDDFT periodic with Hybrid functional

--
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<mailto:cp2k+uns...@googlegroups.com>.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/2057888c-c55f-46f1-877e-88d130612009n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/2057888c-c55f-46f1-877e-88d130612009n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[oandr...@gmail.com]

Dear. Prof Jutter, 

Thank you for your email , 

If I may , I did a calculation without the description of XC in the PROPERTIES section in a system that is a singlet and I compared with the RKS and UKS solutions , 

However I have some differences in the first excitations energies. 

I was wondering if this difference can be as a result of the KERNEL ,  ( though I expected differences between RKS and UKS , I am not sure If in this case is correct_

 R-TDDFPT states of multiplicity 1
 Transition dipoles calculated using velocity formulation

         State    Excitation        Transition dipole (a.u.)        Oscillator
         number   energy (eV)       x           y           z     strength (a.u.)
         ------------------------------------------------------------------------
 TDDFPT|      1       1.28794  -4.0827E-01 -2.5202E-01 -1.7990E-01   8.28480E-03
 TDDFPT|      2       1.33263   5.2202E-01 -8.8606E-01  3.4546E-03   3.45302E-02
 TDDFPT|      3       1.54642  -5.2481E-01  8.9282E-01  9.5216E-04   4.06359E-02
 TDDFPT|      4       1.56965   1.1752E+00  7.0660E-01 -3.0917E-02   7.23448E-02
 TDDFPT|      5       1.63983  -4.6491E-01  7.8123E-01  7.6389E-04   3.32030E-02
 TDDFPT|      6       1.74362   8.1594E-01  4.7197E-01  8.3036E-02   3.82494E-02
 TDDFPT|      7       1.83057  -2.1331E-01  3.3635E-01  2.2744E-04   7.11434E-03
 TDDFPT|      8       1.87648  -6.5168E-01 -3.7868E-01  2.2878E-02   2.61403E-02
 TDDFPT|      9       1.88707   3.5067E-01  8.0384E-01 -3.9408E-03   3.55596E-02
 TDDFPT|     10       1.88755   1.0765E+00  1.9611E-01 -4.4855E-03   5.53681E-02

 U-TDDFPT states of multiplicity 1
 Transition dipoles calculated using velocity formulation

         State    Excitation        Transition dipole (a.u.)        Oscillator
         number   energy (eV)       x           y           z     strength (a.u.)
         ------------------------------------------------------------------------
 TDDFPT|      1       0.84089   6.5188E-01  1.1321E+00  8.8625E-04   3.51567E-02
 TDDFPT|      2       0.93313   6.3627E-01  3.6431E-01  1.7204E-01   1.29659E-02
 TDDFPT|      3       1.16569   1.0576E+00  6.2774E-01  1.3961E-01   4.37563E-02
 TDDFPT|      4       1.27747   4.3857E-01  7.2817E-01  3.5816E-03   2.26149E-02
 TDDFPT|      5       1.28776   2.8845E-01  1.7849E-01  1.2716E-01   4.14021E-03
 TDDFPT|      6       1.33256   3.6916E-01  6.2625E-01  2.5210E-03   1.72533E-02
 TDDFPT|      7       1.42774   2.1708E-01  3.7900E-01  5.2884E-03   6.67373E-03
 TDDFPT|      8       1.44971   6.4799E-02  1.8375E-02  1.9259E-01   1.47852E-03
 TDDFPT|      9       1.54640   3.7002E-01  6.3152E-01  6.5700E-04   2.02970E-02
 TDDFPT|     10       1.56956   8.3101E-01  4.9878E-01  2.1858E-02   3.61402E-02

Thank you again for the time, 

Andres Ortega-Guerrreo

nanotech@surface Empa 

[Jürg Hutter]

Hi

RKS -> default is singlet excitations, no triplets
UKS -> all states includeing quasi 'singlets' and 'triplets'

check the alignment of excitation energies.

regards
JH

________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of oandr...@gmail.com <oandr...@gmail.com>

Sent: Friday, March 7, 2025 10:57 AM
To: cp2k
Subject: Re: [CP2K:21269] TDDFT periodic with Hybrid functional

To view this discussion visit https://groups.google.com/d/msgid/cp2k/50430063-731c-4d68-83d8-f590c3ecdba9n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/50430063-731c-4d68-83d8-f590c3ecdba9n%40googlegroups.com?utm_medium=email&utm_source=footer>.