[CPMD-list] TD_METHOD_A SAOP again
Axel Kohlmeyer
akohlmey at cmm.chem.upenn.edu
Tue Jul 17 22:28:48 CEST 2007
On Tue, 17 Jul 2007 enrico.tapavicza at epfl.ch wrote:
ET> Dear Mailinglist,
ET>
ET> I am using TD_METHOD_A SAOP in
ET> a.) molecular dynamics
ET> b.) singlepoint electronic spectra calculation.
ET> I do not get the same excitation energies if I use b.) for the geometries
ET> generated by a.).
ET> Furthermore the excitation energies using a.) are very spiky, but
ET> ground state energies agree in both cases:
ET> E_0 (b) S1 (b) E_0 (a) S1 (a)
ET> -29.659240 6.854 -29.659240 7.87223
ET> -29.659088 6.852 -29.659088 7.99978
ET> -29.658837 6.847 -29.658837 7.38811
ET> -29.658487 6.842 -29.658487 7.89238
ET> -29.658030 6.837 -29.658030 7.98116
ET> -29.657460 6.831 -29.657460 8.41112
ET> -29.656775 6.824 -29.656775 6.81186
ET> -29.655968 6.817 -29.655968 8.39118
ET> -29.655039 6.810 -29.655039 7.88497
ET> -29.653972 6.801 -29.653972 7.79770
ET> What is the problem and how can it be fixed?
as state before, i have next to no practical knowledge
on TDDFT, but did you check on basis set convergence?
70ry is at the very lower end of convergence for
oxygen and it works for CP-dynamics mostly because
of error cancellation. however, some papers show,
that about 85ry is a more reasonably choice...
and this is for the ground state. for excited states
i would expect this to be much more sensitive to
basis set completeness...
please can some of the real TDDFT experts here
confirm or disconfirm this. thanks.
cheers,
axel.
ET>
ET> Thanks for help!
ET> Enrico
ET>
ET> Here is input for a.):
ET> &CPMD
ET> MOLECULAR DYNAMICS BO
ET> TDDFT
ET> RESTART WAVEFUNCTION COORDINATES LINRES LATEST GEOFILE
ET> TIMESTEP
ET> 10
ET> TEMPERATURE
ET> 300
ET> MAXSTEP
ET> 10
ET> CONVERGENCE ORBITALS
ET> 1.E-6
ET> TRAJECTORY XYZ
ET> &END
ET>
ET> &TDDFT
ET> TAMM-DANCOFF
ET> STATES SINGLET
ET> 6
ET> FORCE STATE
ET> 1
ET> DAVIDSON RDIIS
ET> 20 3 1.D-9
ET> DAVIDSON PARAMETER
ET> 2000 1.D-7 50
ET> TD_METHOD_A SAOP
ET> &END
ET>
ET> &LINRES
ET> CONVERGENCE
ET> 1.D-5
ET> MAXSTEP
ET> 1000
ET> HTHRS
ET> 1.0
ET> DIFF FORMULA
ET> 4
ET> oPTIMIZER PCG
ET> tHAUTO
ET> 1.D-1 1.D-3
ET> &END
ET>
ET> &DFT
ET> NEWCODE
ET> FUNCTIONAL PBE
ET> &END
ET>
ET> &SYSTEM
ET> ANGSTROM
ET> SYMMETRY
ET> 0
ET> POISSON TUCKERMAN
ET> CELL ABSOLUTE
ET> 9.0 9.0 9.0 0 0 0
ET> CUTOFF
ET> 70.0
ET> &END
ET> &ATOMS
ET> *C_MT_PBE KLEINMAN-BYLANDER
ET> LMAX=P
ET> 2
ET> 4.264730590306 5.000000355901 5.418154071495
ET> 5.735269561836 5.000000277565 5.418153972768
ET> *O_MT_PBE KLEINMAN-BYLANDER
ET> LMAX=P
ET> 1
ET> 4.999999604830 4.999992782686 4.173841430278
ET> *H_MT_PBE KLEINMAN-BYLANDER
ET> LMAX=S
ET> 4
ET> 3.723365639105 5.928941831162 5.633801690656
ET> 6.276638331045 4.071061532101 5.633796460623
ET> 3.723362039879 4.071061615679 5.633796577089
ET> 6.276634232991 5.928941604892 5.633802270584
ET> &END
ET>
ET> Here is input for b.):
ET> &CPMD
ET> ELECTRONIC SPECTRA
ET> rESTART WAVEFUNCTION COORDINATES LATEST
ET> CONVERGENCE ORBITALS
ET> 1.E-6
ET> CENTER MOLECULE
ET> &END
ET>
ET> &TDDFT
ET> TAMM-DANCOFF
ET> STATES SINGLET
ET> 6
ET> DAVIDSON RDIIS
ET> 20 3 1.D-9
ET> DAVIDSON PARAMETER
ET> 2000 1.D-6 50
ET> TD_METHOD_A SAOP
ET> &END
ET>
ET> &LINRES
ET> CONVERGENCE
ET> 1.D-5
ET> MAXSTEP
ET> 1000
ET> HTHRS
ET> 1.0
ET> DIFF FORMULA
ET> 4
ET> oPTIMIZER PCG
ET> tHAUTO
ET> 1.D-1 1.D-3
ET> &END
ET>
ET> &DFT
ET> NEWCODE
ET> FUNCTIONAL PBE
ET> &END
ET>
ET> &SYSTEM
ET> ANGSTROM
ET> SYMMETRY
ET> 0
ET> POISSON TUCKERMAN
ET> CELL ABSOLUTE
ET> 9.0 9.0 9.0 0 0 0
ET> CUTOFF
ET> 70.0
ET> &END
ET>
ET>
ET> &ATOMS
ET> *C_MT_PBE KLEINMAN-BYLANDER
ET> LMAX=P
ET> 2
ET> 4.264730590306 5.000000355901 5.418154071495
ET> 5.735269561836 5.000000277565 5.418153972768
ET> *O_MT_PBE KLEINMAN-BYLANDER
ET> LMAX=P
ET> 1
ET> 4.999999604830 4.999992782686 4.173841430278
ET> *H_MT_PBE KLEINMAN-BYLANDER
ET> LMAX=S
ET> 4
ET> 3.723365639105 5.928941831162 5.633801690656
ET> 6.276638331045 4.071061532101 5.633796460623
ET> 3.723362039879 4.071061615679 5.633796577089
ET> 6.276634232991 5.928941604892 5.633802270584
ET> &END
ET>
ET>
ET> =============================================================
ET> Enrico Tapavicza
ET> Laboratory of Computational
ET> Chemistry and Biochemistry
ET> BCH 4118 EPF Lausanne
ET> CH-1015 Lausanne
ET> Tel:++41-(0)21-693-0326
ET> Mobil:++41-(0)76-2063421
ET> Fax: ++41-(0)21-693-0320
ET> Email: enrico.tapavicza[at]epfl.ch
ET> =============================================================
ET> _______________________________________________
ET> CPMD-list mailing list
ET> CPMD-list at cpmd.org
ET> http://cpmd.org/mailman/listinfo/cpmd-list
ET>
--
=======================================================================
Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu
Center for Molecular Modeling -- University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.
More information about the CPMD-list
mailing list