It is difficult to give a reasonable answer without the whole input file, but I’d suggest to verify that the charge, i.e. the deprotonation of the porphyrine is correct. In addition I can’t understand why the BS for Zn is doubled at the end of the input file.

Hi Massimiliano Arca thanks for your reply, the doubled BS at the end was a mistake when I copy/paste the input. But here is the full input that I am using.

The problem derives from the adoption of the lanl2dz bs for Zn. In fact, by using the 6-31g(d) bs for all atomic species, the calculation proceeds smoothly. You might try two solutions. First, you may want to avoid a bs with ECP for a light atomic species such as Zn, and therefore another set, such as 6-31g(d) itself, or a def2 bs. Second, if you prefer to keep the LANL2DZ BS you must specify the ECPs:

C H N O 0

6-31g(d)

****

Zn 0

LANL2DZ

****

ZN 0

LANL2DZ

You can also try a different formulation of the lanl2dz bs (https://www.basissetexchange.org/basis/lanl2dz/format/gaussian94/?version=0&elements=30). In all cases, remember to add pseudo=read in the command line of the input file.

Hallo, Paola!

If you checked your output, you surely found no pseudopotential data for Zn.

You must replace keywords B3LYP/Gen by

B3LYP/GenECP

or

B3LYP/Gen Pseudo=Read

Best wishes

mb

Martin Breza, This suggestion worked perfectly! Thank you so much!