One of the most successful posts this blog has ever published was on certain nuances of the solvation calculations on PCM in G03. However there are some differences in the SCRF modules between G09 and G03 and I here present some of them as well as some tips to work with the new version.

The SCFVAC keyword used to calculate the Gibbs Solvation Energy change is no longer available. It is now replaced by DoVacuum which should be included in the SCRF options as SCRF=(DoVacuum,etc.). However, the absolute solvation energy now requires a gas-phase optimization along with a frequency calculation followed by the same calculations with the SCRF=SMD option in the desired solvent and with the appropriate variables.

Gaussian 03 used to calculate and report non-electrostatic contributions to the solvation energy, however they were not included in the total energy nor during optimization procedures. These non-electrosatic interactions are no longer calculated in the default. In order to include these terms during the SCF procedure, and to have them reported separately, the SCRF=SMD option should be used.

My previous post on PCM mentioned the usage of the options OFac=0.8 & RMin=0.5 as part of the additional input. These ‘magic numbers’ (I hate the term) were used to modify the way by which the overlapping spheres were treated in order to create the surface which in turn defined the cavity. G09 uses a new algorithm to make the overlaps generate a smoother surface. I recommend to use the default values before including ‘magic parameters’.

All the default values which G03 used can be retrieved with the G03Defaults keyword, but it is strongly suggested to use it only for comparison with calculations previously done with the older version.

As with some other so-called ‘white papers’ this post will be further improved as more information arises during my own calculations. Thanks for reading! Please comment/like/share this post, as well as others in the blog, if you found useful the information within.