njackson7643
commented
4 years ago Hi - Sorry for the slow response and thanks for the interest!
The SMD calculation defaults to a IEFPCM model calculation - so it is just a PCM solvation energy with the non-electrostatic terms derived from SMD radii. The relevant line in Gaussian we use if you want to reproduce it for a given molecular structure is '#N B3LYP/GEN SP SCF=Tight SCRF=(SMD,DoVacuum) Symmetry=None' in Gaussian 2016 (GEN is just set to 6-31G** and all other parameters are default). Note that the defaults for SCRF changed in the '09 version, so be careful of the versioning. We will be updating the details of the solvation calculation protocol in a revised upcoming version of the manuscript to clear up this point - I've attached a representative output file from the QC calculations if you want more details. I don't have any good answers about the lack of correlation with exp data that you reference, particularly without further details about the exp. However, we were primarily interested in the solvation energies as a qualitative characterization of the 3D intermolecular interaction environment that might be relevant to melting point prediction and it appears that, regardless of accuracy, they do convey some degree of information. Hope this helps.
Best,
Nick
pythonpanda2
Hi,
Nice work! Many very interesting ideas from the paper!
Since you calculated solvation energy for those compounds, I compared the calculated values to experimental values for the compounds with available data, and I found that there is no clear correlation between the calculated and experimental values. Can you provide more details about the SMD calculation?
Thanks!