How to use a solvent mixture in gaussian ?

mekelleche

Hi everyone,
I would like to study the solvent effect of an organic reaction in a
solvent mixture, Methanol:THF (80:20) using gaussian 09 software, but gaussian doesn't support it.
Please, i need your help and suggestion.

冰释之川

You can customize eps and epsinf two parameters (eps and epsinf denote static dielectric constant and dynamic dielectric constant, respectively) after the atom coordinate section of the Gaussian input file (a blank line is needed to separate the atom coordinate section and the eps and epsinf parameters).
A solvent mixture can be thus implemented by using the volume-weighted eps and epsinf values of all involved solvents.

Note that the epsinf parameter is not easy to get, but for nonpolar solvents, it often approximates to the square of refractive index or the eps parameter. As for polar solvents, the epsinf value is usually smaller than the corresponding eps value by about an order of magnitude.

In your case, the eps and epsinf values of the mixed solvent equal to 0.8*eps(methanol)+0.2*eps(THF) and 0.8*epsinf(methanol)+0.2*epsinf(THF), respectively (provided that methanol:THF=80:20 is in volume proportion), where eps(methanol)=32.613 and eps(THF)=7.4257 according to the manual of Gaussian09 (http://sobereva.com/g09/k_scrf.htm).