Dear All,
I have been trying to perform TDDFT calculation for the Ga-phthalocyanine complex, and this calculation has been terminated with the following error message:
Largest valence mixing into a core orbital is 5.94D-01
Largest core mixing into a valence orbital is 3.87D-01
Excessive mixing of frozen core and valence orbitals.
Error termination via Lnk1e in /opt/g16/l801.exe at Sat Jul 24 14:46:19 2021.
Below there are data of my calculations:
the system: C68H48ClGaN8O4
basis set: 6-31G*
method: B3LYP or wB97XD
implicit solvent: THF
# of basis functions: 1345
symmetry: C1
charge and spin: 0 1
software: Gaussian 16, revision B.01
What could be the reason for this failure? Unfortunately, I could not find the answer by googling for it.
Thank you very much in advance!
Best regards,
Aleksey.
Dear Aleksey E. Kuznetsov,
Are you using pseudopotential for Ga?
Could you please share your input file? Probably the first line (# ...) will be enough to have an idea.
Let's see if we can solve the problem!
Best regards,
Fernando
This is a well-known issue in calculations with alkali metals described e.g. by Petrie et coauthors in J. Phys. Chem. A 102, 6138 (1998) and by Rassolov and coauthors in Chemical Physics Letters 350, 573 (2001) and maybe holds too for Ga.
The problem is that the semi-core orbitals of the alkali could have orbital energies are higher than the valence orbitals of other species, which means problems in the active space of the TD-DFT calculation. This could be solved increasing the size of the active space such that all valence orbitals are included therein. Was suggested, despite the computational cost, to use td=full.@
Dear Massimo Ottonelli , thank you very much for your advice! I used td(full) option, and it indeed worked.
With kind regards,
Aleksey.
Dear Fernando Aguilar-Galindo , thank you for your response, too! I used full basis set for Ga as well, and after I used the TD(full) option, everything worked properly.
With kind regards,
Aleksey.
Dear Aleksey, I'm working in a project of phytolocyanin photosensitizers with Zn.
The value of the experimental S0 to S1 transition is 681 nm. To choose a better wavefunction that represented the electronic structure, I did a benchmark study between DFTs/basis set.
The best DFT/basis set combination that generated the exact wavelength (681nm) using TD-DFT was BhandHLYP/6-311G(2d,2p). In second place were BHandHLYP/Def2TZVP and Def2DZVP. Try these combinations, it might help you to solve your problem.
Best Regards
- Badges
- Science topic