Of course you can add a dielectric constant of a solvent but it is dangerous, the problem is that dielectric constant is one of many parameters that characterize a solvent. For example methanol has a dielectric constant lower than acetonitrile, but it dissociate acids easy than acetonitrile, because it is a protic solvent while acetonitrile is aprotic. The best solution is to find as close as possible solvent. Just as If you want to mimic lipid environment in human body and provided that lipids are esters with long hydrophobic chain, the solution is to use pentyletanoate, the largest ester included in Gaussian parametrization. However many authors use benzene which has approximately the same dielectric constant but lack of oxygen that can be a H bond electron pair donor, however it works.

You can put the dielectric constant of the solvent in the additional keyword section, i.e. after the geometry specification (don't forget to leave a blank line between the last line of the geometry specification and additional keyword section). An abridged example from http://www.gaussian.com/g_tech/g_ur/k_scrf.htm  is shown below:

# B3LYP/6-31G(d) 5D SCRF(SMD,Solvent=Generic,Read)

Water, solvation by methanol, re-defined as generic solvent.

0 1

O

H,1,0.94

H,1,0.94,2,104.5

stoichiometry=C1H4O1 Input section for PCM keywords

solventname=methanol

eps=32.63

epsinf=1.758

…

Blank line terminates PCM input

Of course you can add a dielectric constant of a solvent but it is dangerous, the problem is that dielectric constant is one of many parameters that characterize a solvent. For example methanol has a dielectric constant lower than acetonitrile, but it dissociate acids easy than acetonitrile, because it is a protic solvent while acetonitrile is aprotic. The best solution is to find as close as possible solvent. Just as If you want to mimic lipid environment in human body and provided that lipids are esters with long hydrophobic chain, the solution is to use pentyletanoate, the largest ester included in Gaussian parametrization. However many authors use benzene which has approximately the same dielectric constant but lack of oxygen that can be a H bond electron pair donor, however it works.

Mr. Ojo,

Please pay attention ti a discussion:

https://www.researchgate.net/post/Do_I_have_to_manually_enter_dielectric_constants_for_PCM_solvents_in_DFT_calculations_in_Gaussian_at_elevated_temperature

and relating attachments. there is shown illustratively by an open Gaussian window how to do this.