Modeling Solvation Effects
I recommend using a simple, but rather accurate, approach--the Polarized Continuum Model (PCM). PCM has three key components: (1) Instead of treating individual solvent molecules, PCM treats the solvent as a "continuum"--a fluid of a certain polarity. This polarity is quantified by a property known as a dielectric constant. (Think of the dielectric constant as a macroscopic version of a solvent molecule's individual dipole moment.) (2) The electron density on the solute is allowed to distort (or become polarized) in response to the polarity of the solvent. (3) The (rather small) energetic cost of opening a "pocket" in the solvent to let the solute dissolve is calculated by estimating the size a given solute takes up in solution.
SCRF=(PCM,SOLVENT=YYY)
where YYY=H2O, CH2CL2, etc. (See the instructions on the Gaussian 03 web page for a complete list of possible solvents.) Also, you can just add
SCRF=PCM
Gaussian 03 will assume the solvent is water.
(Polarized solute)-Solvent (kcal/mol) = -80.85
Total non electrostatic (kcal/mol) = 2.43
The first line takes into account components (1) and (2) discussed above, and the second line takes into account component (3) discussed above. Note that the first energy will always be negative--any molecule or ion will be more stable when dissolved in a solvent--and the second energy will always be positive. Both energies should be added to the energy you have already determined from your previous geometry optimization/vibrational frequency calculation--either a zero-point-corrected electronic energy or an enthalpy.
Please let me know if you have any questions!