[CP2K:2818] Re: TDDFPT calculation
Juerg Hutter
hut... at pci.uzh.ch
Thu Aug 26 08:38:06 UTC 2010
Hi
thanks for the information.
Please note, that the implementation of TDDFT in cp2k
does not restrict the space of orbitals for the excitation.
However, in order to generate an initial guess a certain
number of occupied and unoccupied orbitals are chosen.
I assume that in your case the space of unoccupied
orbitals was limited to the LUMO and the diagonalization
later was not able to find other states from these starting states.
regards
Juerg
----------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut... at pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
----------------------------------------------------------
On Wed, 25 Aug 2010, xujroman wrote:
> Dear Juerg Hutter,
>
> thank you for your answer.
>
> However, I have found the problem in my cp2k calculation.
>
> I have used the following setting in the DFT-PRINT section.
>
> .
> .
> .
> &PRINT
> &MO_CUBES
> FILENAME ./MOTEST
> STRIDE 1 1 1
> NLUMO 1
> NHOMO 1
> &END MO_CUBES
> &MO
> EIGENVALUES .TRUE.
> OCCUPATION_NUMBERS .TRUE.
> &END MO
> &END PRINT
> .
> .
> .
>
> This leads to the calculation of only the first unoccupied molecular
> orbital. Therefore, I obtained only transitions into this state.
> Changing the above section into
>
> .
> .
> .
> &PRINT
> &MO_CUBES
> WRITE_CUBE .FALSE.
> NLUMO 10
> &END MO_CUBES
> &MO
> EIGENVALUES .TRUE.
> OCCUPATION_NUMBERS .TRUE.
> &END MO
> &END PRINT
> .
> .
> .
>
> solved the problem.
>
> However, I found it not intuitive to specify the number of used
> unoccupied orbitals of the TDDFT calculation in the section FORCE_EVAL-
>> DFT-> PRINT-> MO_CUBES .
> Therefore, I suggest for the next cp2k release a new keyword in the
> TDDFPT section which specifies the number of LUMOS or at least a hint
> in the manual where the number of LUMOS have to be set.
>
>
> Cheers,
> Roman
>
>
>
>
> On Aug 24, 12:56 pm, Juerg Hutter <hut... at pci.uzh.ch> wrote:
>> Hi
>>
>> cp2k uses the Tamm-Dancoff approximation to TDDFT. This is known
>> to give excitations that are often very strongly localized (in MO
>> space). Do you have evidence that the result you get is not correct?
>>
>> regards
>>
>> Juerg
>>
>>> Dear cp2k users,
>>
>>> I have tried to to a TDDFT calculation with cp2k.
>>> The input file was as follows:
>>
>>> &FORCE_EVAL
>>> METHOD Quickstep
>>> STRESS_TENSOR ANALYTICAL
>>> &DFT
>>> EXCITATIONS tddfpt
>>> &TDDFPT
>>> MAX_KV 60
>>> NEV 8
>>> OE_CORR SAOP
>>> REORTHO 2
>>> RES_ETYPE SINGLETS
>>> RESTARTS 5
>>> &XC
>>> &XC_FUNCTIONAL PBE
>>> &END XC_FUNCTIONAL
>>> &END XC
>>> &END TDDFPT
>>> &PRINT
>>> &MO_CUBES
>>> FILENAME ./MOTEST
>>> STRIDE 1 1 1
>>> NLUMO 1
>>> NHOMO 1
>>> &END MO_CUBES
>>> &MO
>>> EIGENVALUES .TRUE.
>>> OCCUPATION_NUMBERS .TRUE.
>>> &END MO
>>> &END PRINT
>>> &POISSON
>>> PERIODIC NONE
>>> POISSON_SOLVER MULTIPOLE
>>> &MULTIPOLE
>>> ANALYTICAL_GTERM .TRUE.
>>> RCUT 8.0
>>> &END MULTIPOLE
>>> &END POISSON
>>> &MGRID
>>> CUTOFF $i
>>> REL_CUTOFF 100
>>> &END MGRID &QS
>>> WF_INTERPOLATION PS
>>> EXTRAPOLATION_ORDER 3
>>> &END QS
>>> &SCF
>>> SCF_GUESS ATOMIC
>>> ADDED_MOS 11
>>> MAX_SCF 500
>>> &END SCF
>>> &XC
>>> &XC_FUNCTIONAL PBE
>>> &END XC_FUNCTIONAL
>>> &END XC
>>> &END DFT
>>> &SUBSYS
>>> &CELL
>>> PERIODIC NONE
>>> A 16 0 0
>>> B 0 16 0
>>> C 0 0 16
>>> &END CELL
>>> &COORD
>>> C 0.0000015548 1.3995049822 0.0000112361
>>> C 1.2105571712 0.6993498626 -0.0000011831
>>> C 1.2105625355 -0.6993566874 -0.0000046545
>>> C 0.0000152421 -1.3995045587 -0.0000022992
>>> C -1.2105874557 -0.6993739468 0.0000097107
>>> C -1.2105811121 0.6993813716 0.0000137609
>>> F -0.0000016621 2.7455352497 -0.0000018655
>>> F -2.3775341687 1.3700043040 0.0000017066
>>> F -2.3775319035 -1.3700126878 -0.0000055826
>>> F -0.0000018753 -2.7455346120 -0.0000011244
>>> F 2.3775445180 -1.3700080734 0.0000004072
>>> F 2.3775535520 1.3700123130 -0.0000037231
>>> &END COORD
>>> &KIND C
>>> BASIS_SET DZVP-MOLOPT-SR-GTH
>>> POTENTIAL GTH-PBE-q4
>>> &END KIND
>>> &KIND F
>>> BASIS_SET DZVP-MOLOPT-SR-GTH
>>> POTENTIAL GTH-PBE-q7
>>> &END KIND
>>> &END SUBSYS
>>> &END FORCE_EVAL
>>> &GLOBAL
>>> PROJECT SINGLE-C6F6
>>> RUN_TYPE SPECTRA
>>> PRINT_LEVEL MEDIUM
>>> &END GLOBAL
>>> &MOTION
>>> &GEO_OPT
>>> OPTIMIZER CG
>>> &CG
>>> &LINE_SEARCH
>>> TYPE 2PNT
>>> &END LINE_SEARCH
>>> &END CG
>>> &END GEO_OPT
>>> &END MOTION
>>
>>> As output for the excited states I obtained the following:
>>
>>> excited state : 1 5.205378
>>
>>> 33 -> 34 0.693 0.480
>>> 32 -> 34 0.082 0.487
>>> 20 -> 34 0.109 0.499
>>
>>> excited state : 2 7.547600
>>
>>> 33 -> 34 0.108 0.012
>>> 32 -> 34 0.973 0.959
>>> 21 -> 34 0.149 0.982
>>
>>> excited state : 3 7.785291
>>
>>> 33 -> 34 0.671 0.450
>>> 32 -> 34 0.070 0.455
>>> 31 -> 34 0.066 0.459
>>> 20 -> 34 0.065 0.464
>>
>>> excited state : 4 8.544974
>>
>>> 31 -> 34 0.962 0.928
>>> 19 -> 34 0.228 0.980
>>
>>> excited state : 5 8.663582
>>
>>> 30 -> 34 0.977 0.954
>>> 13 -> 34 0.207 0.997
>>
>>> excited state : 6 9.602348
>>
>>> 29 -> 34 0.059 0.004
>>> 28 -> 34 0.986 0.976
>>> 18 -> 34 0.084 0.984
>>> 11 -> 34 0.114 0.997
>>
>>> excited state : 7 9.771548
>>
>>> 29 -> 34 0.983 0.967
>>> 28 -> 34 0.060 0.971
>>> 17 -> 34 0.102 0.981
>>> 10 -> 34 0.126 0.997
>>
>>> excited state : 8 10.868760
>>
>>> 27 -> 34 1.000 0.999
>>
>>> TDDFPT : CheckSum = 0.899094E+00
>>
>>> Now my question are:
>>
>>> Why there are only excitations to the first unoccupied orbital (34)
>>> and no contributions higher orbitals (e.g. 35, 36 etc.) ?
>>
>>> Do I have to specify the number of unoccupied orbitals used in the
>>> TDDFT section?
>>
>>> If, yes, where do have to specify this?
>>
>>> Sincerely,
>>> Roman
>>
>>> --
>>> You received this message because you are subscribed to the Google Groups "cp2k" group.
>>> To post to this group, send email to cp... at googlegroups.com.
>>> To unsubscribe from this group, send email to cp2k+uns... at googlegroups.com.
>>> For more options, visit this group athttp://groups.google.com/group/cp2k?hl=en.
>>
>>
>
> --
> You received this message because you are subscribed to the Google Groups "cp2k" group.
> To post to this group, send email to cp... at googlegroups.com.
> To unsubscribe from this group, send email to cp2k+uns... at googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/cp2k?hl=en.
>
>
More information about the CP2K-user
mailing list