Can I do Two-photon absorption using a solvent (COSMO)?

[dpleft]

Hello all,
I a doing two-photon absorption calculations and I am interested in using a solvent. The particular solvent I am trying to use in the followed input control file is DMSO.
This is my control file,

Code: [Select]

$lastdiag
$rpacor    400
$maxcor    500 MiB  per_core
$dft
 functional cam-b3lyp
 gridsize m4
$soes all    1
$scfiterlimit    100
$coord    file=coord
$scfconv    6
$grad    file=gradient
$scfinstab twophoton rpas
$cosmo_atoms
#radii in Angstrom units
o 23-24 \
radius=1.72
s 25 \
radius=2.16
n 21-22 \
radius=1.83
h 26-43 \
radius=1.3
c 1-20 \
radius=2.0
$optimize
 internal off
 redundant off
 cartesian on
 global off
$symmetry    c1
$basis    file=basis
$cosmo_out    file=job_SP_18.cosmo
$scfmo   file=mos
$cosmo
 rsolv=1.3
 refind=1.3
 use_contcav
$atoms
    basis =aug-cc-pVDZ
$scfdamp     start=0.700 step=0.050 min=0.100
$operating system    unix
$closed shells
 a       1-92                                   ( 2 )
$scfdump
$scfdiis
$scforbitalshift  automatic=.1
$energy    file=energy
$rundimensions
   natoms=43
$end
The run terminates successfully and provides me with reasonable results, but I want to confirm that this actually works well and it is implementation should lead to that. Moreover, for the solvent and selection of solvation parameters, I am doing them manually, does that make sense?

I appreciate answers into this.

[dpleft]

Anyone with a follow regarding this?

[antti_karttunen]

Hi,

I never tried to use COSMO with TPA, but here are some comments:

1. For "solvent choice", the most crucial parameter is epsilon (dielectric constant). The default is infinity, which is fine for water. For weak dielectrics you may want to check the effect of this (see chapter 19.2. in the manual)
2. For excited state calculations, the solvent response can be divided in so-called slow and fast terms (manual Chapter 19.2.3). You may want to check how your results look like if you just use COSMO for the ground state calculation (ridft) and take it out for escf. In any case, the output should give you more details on the use of COSMO, please check the output from escf.

Best,
Antti

[dpleft]

Thank you for your answer. I will look more into it and check these parameters and learn more about COSMO.