Hi Youssef Nada ,

calculation of the deltaG(solv) is a bit tricky and involved but for the start.

Assume you have a structure in gas phase and the corresponding gibbs free energy using a frequency calculation. Then you would like to know the free energy within a solvent, that is the total energy + the thermodynamic correction as you had in vaccum/gas.

In theory, the molecule could adopt to the solvent and change its geometry and thence its vibrational modes. If it does not, then the only change comes from the difference of the electronic energy in the solvent, the geometry and thermdynamics are the same. So only a single point in gas and solvent is necessary, both using the same level of theory. The only changes comes from the change in the continuum.

But there is one very important thing the keep in mind, even before dealing with the solvation model itself. The reference state in of the ideal gas in gas phase is different from the ideal solution in solution.

You have to deal with this, I will give a link with some practise, but here is the full explanation:

Essentials of Computational Chemistry: Theories and Models, 2nd Ed., ISBN-13 978-0470091821.

Now for the solvation models. Different programs have different approaches, which might be documented with different level of detail. Some give only the electronic energy in the solvent, some give already a term for the energy of solution, some are heavily parametrised ( to recover experimental energies of solvation), some are not.

If you do not want to deal with the stuff, you can just follow the tutorial for orca:

https://www.orcasoftware.de/tutorials_orca/prop/CPCM.html and either adopt it for gaussian or use orca.

https://github.com/grimme-lab/xtb/releases/tag/v6.6.0 can be used to calculate free energy of solvation

The commercial software CosmoTherm (Cosmologic) can be used as well, it however follows a different approach. There is a free alternative:

https://github.com/TUHH-TVT

I hope this does not confuse further ...

Kind regards,

Conrad

Plus, In this paper they used Gaussian 98, and I am using Gaussian 16 so will this make a difference?

Yes. Below you can see the output of Gaussian98 when calculated taking into account the influence of the solvent.

----------------------------------------------------                                                                                                                                   -------------- VARIATIONAL COSMO RESULTS -------------                                                                                                                                       ------------------------------------------------------                                                                                                                                                  (a.u.) =  -2692.899759                                                                                                                                               (a.u.) =  -2692.919123                                                                                                                                               (a.u.) =  -2692.921033                                                                                                                                                    (a.u.) =  -2692.897474                                                                                                                                               (a.u.) =  -2692.921068                                                                                                                                            Total free energy in sol.                                                                                                                                                                    (with non electrost.terms) (a.u.) =  -2692.909287                                                                                                                                         ------------------------------------------------------                                                                                                                                          (Unpol.Solute)-Solvent (kcal/mol) =    -12.15                                                                                                                                                (Polar.Solute)-Solvent (kcal/mol) =    -14.81                                                                                                                                                Solute Polarization    (kcal/mol) =      1.43                                                                                                                                                Total Electrostatic    (kcal/mol) =    -13.37                                                                                                                                             ------------------------------------------------------                                                                                                                                          Cavitation energy      (kcal/mol) =     20.28                                                                                                                                                Dispersion energy      (kcal/mol) =    -15.86                                                                                                                                                Repulsion energy       (kcal/mol) =      2.97                                                                                                                                                Total non electr.      (kcal/mol) =      7.39                                                                                                                                             ------------------------------------------------------                                                                                                                                          DeltaG (solv)          (kcal/mol) =     -5.98                                                                                                                                             ------------------------------------------------------

Unfortunately, it is typical for Gaussian Inc. to change the code with each version so that some results obtained in the previous version of the program are not reproduced without additional adjustments through the IOP options. Or, as in the case of deltaG(sol), it is necessary to calculate differently. When using later versions of Gaussian, we evaluated this value as the difference between the total energies in the gas phase and in solution + Gcorr for solution

Hello Alexey N. Masliy ,

Thank you very much for your reply. Are you aware of how much of a difference should it make? I even copied their geometries in the supplementary information and did the SPC , and there is a 1 Hartree value difference. I wonder where the difference is. The only correction I make is the gas from (1 atm) to 1 mol/L which is 1.89 kcal/mol. Is there something I should make?

In my opinion, such a significant difference is due to the fact that you lost hydrogen or incorrectly indicated the charge of the particle. There is no other way I can explain such a significant difference. Or you need to look at the gjf files in the original work and yours.