PCM- Psi4 vs. Gaussian


I have a problem with setting Psi4 so that the EA results (for HOMO and LUMO) are the same as in Gaussian for the HF method.
Additionally, for ordinary HF calculations we have full compliance, while with PCM there is a problem.

How should Psi4 be set to get the same results as in Gaussian?
Below I present inputs for Psi4 and Gaussian:

For Psi4:

molecule NH3 {
symmetry c1
N 0.0000000 0.0000000 0.2851841
H -0.4697490 0.8136292 -0.0950614
H -0.4697490 -0.8136292 -0.0950614
H 0.9394980 0.0000000 -0.0950614

set {
basis def2-TZVPP
scf_type pk
pcm true
pcm_scf_type total

pcm = {
Units = Angstrom
Medium {
SolverType = IEFPCM
Solvent = Water

Cavity {
RadiiSet = UFF
Type = GePol
Scaling = False
Area = 0.3
Mode = Implicit


For Gaussian:

#p HF/Def2TZVPP SCF=(MaxCycle=100) SCRF=(Solvent=Water)


0 1
N 0.0000000 0.0000000 0.2851841
H -0.4697490 0.8136292 -0.0950614
H -0.4697490 -0.8136292 -0.0950614
H 0.9394980 0.0000000 -0.0950614

Thank you in advance for any suggestions or insight you might have for me!


How big of a difference are you seeing?


This is very difficult and the best hope is to get close.
Gaussian uses default features that PSI4/PCMSOLVER does not have.
So you need Gaussian to use the same settings as PCMSOLVER.
Using the same radii and alpha scaling should go a long way to get close.
Check the Additional Input section in the G16 online manual for SCRF.

Compare the PCM output in both programs to see differences and try to find settings that bring them in agreement.

Even G16 cannot exactly reproduce G09 results, to give you an idea how sensitive the algorithms are to small changes.

Total energy:
For Psi4 = -56.2283090368
For Gaussian = -56.2264561257