Viewing MOs with Various Engines

[Hypersphere]

1. When I run a molecular energy calculation with GAMESS 2022.2 for Linux, the output automatically includes molecular orbitals, but this is not the case with other engines, such as Gaussian16 or Psi4 1.7. Would I need to determine the necessary keywords for each engine to get MO output from, say, an energy or geometry optimization calculation?

2. Alternatively, if I am interested in getting MO output, should I just do a separate run and select "molecular orbitals" from the pull-down calculation type menu?

3. When "molecular orbitals" is selected as the calculation type for a given engine, are the MOs calculated by the selected engine, or is the built-in MO function of WebMO being used?

[schmidt]

The easiest way to do this is with an "Molecular Orbital" calculation, which ensures that all of the necessary output is generated. In this sense, GAMESS is an outlier since it ALWAYS (by default) outputs all that is needed anyway; once it is there, WebMO simply parses it.

As a general rule, WebMO doesn't do any calculation. It only parses/displays the results of the underlying computational engine. This is true in the case of the MOs displayed on the view molecule page as well.

One exception to this rule: If you calculate the MOs from the WebMO App (iOS/Android) or directly from the editor (on the Build Molecule page), those MOs and eigenvalues come from WebMO itself, via extended Huckel. They are, of course, extremely approximate.

[jwk]

Here is a strange result using the WebMO app on an ipad v10: Lookup, Import, "crystal violet" gives the Ar3C+ Cl- ion pair. Then calculate orbitals shows the LUMO that looks correct for Ar3C+ ion itself, i.e. with large lobe on the central C. Deleting the Cl- ion, then repeating calculate orbitals gives what looks like LUMO+1, i.e. with no lobe on the central C. If one builds Ar3C+ and adjusts the phenyl ring dihedral angles to about 140 deg, then does calculate orbitals, you get the latter result again.
This is not a problem with Extended Huckel in Gaussian. Using a route section like "#N huckel SP GFINPUT POP=FULL" , both crystal violet cation and the crystal violet ion pair have the expected LUMO distribution with the large central lobe.