Sent to CCL by: Andreas Klamt [klamt*_*cosmologic.de]
Dear Ramon,
you are right: almost all implicit solvation models are parameterized
mainly for water and have a crude scaling for lower dielectric
constants. You cannot expect that the specific details of a solvent as
olive oil is reasonably quantitatively described by such implicit
solvation method. I personally consider the change of te solvent radius
according to the solvent as a very unphysical concept, and I doubt that
there are any good rules how to set that radius.
As far as I know, the only implicit solvation method that can deal with
almost arbitrary solvent and solvents mixtures, even at variable
temperatures without the need for any solvent information (not even the
dielectric constant) is my COSMO-RS method. COSMO-RS describes solutes
and solvents on the same quantum-chemical footing and brings them
together by a statistical thermodynamics of interacting surfaces.
For more details and lots of references see our homepage or my book
"COSMO-RS: From quantum chemistry to fluid phase thermodynamics and drug
design", Elsevier, 2005.
Regards
Andreas
Ramon Crehuet rcsqtc..iiqab.csic.es schrieb:
Sent to CCL by: Ramon Crehuet [rcsqtc.:.iiqab.csic.es]
Dear CCLers,
I would like to do a PCM calculation using GAMESS and this program
needs
some parameters for the solvent. I guess any software needs this
parameters. Ideally I should use olive oil as solvent, but ether has a
similar dielectric constant.
Precisely I don't know where to find the following properties:
* the solvent radius
* the thermal expansion coefficient
* the thermal coefficient of the surface tenstion
* the cavity microscopic coefficient
* solvent numeral density
* distances between the first atoms of each type and the cavity (can
this be just a guess or should this be calculated? And how?)
I have the feeling that usually only water is used as a solvent and
that
although the method in GAMESS is general, the solvent parameters for
other solvents are not trivial and,thus, other solvents are rarely
used.
Has anybody experience with that?
Thanks for your attention,
Ramon Crehuet>