Greetings!

In Gaussian16, you'll need to divide the calculation into fragments, specifying the charge and spin multiplicity for each fragment as well as for the entire molecule. This approach allows you to solve a broken-symmetry structure with fewer difficulties.

You can check the syntax for fragment-based guesses on Gaussian's official website: https://gaussian.com/guess/.

In the "Antiferromagnetic Coupling" section, you'll find an example of a fragment-based guess, which clearly illustrates how to implement it.

I hope this helps!

Lucas.

Thank you Dr. Fachini for your answer

It will be of great help for me.

You can use the following sequence:

%chk=/path to your tmp files folder/O2_Singlet

%nprocshared=32

%mem=40gb

# B3LYP/6-311+G(d) Stable=Opt nosymm scf=xqc

O2 Singlet

0 1

O -0.166353908497 1.084524446627 -0.861089299246

O 0.848567730033 1.712260492530 0.014285777117

--link1--

%chk=/path to your tmp files folder/O2_Singlet

%nprocshared=32

%mem=40gb

# B3LYP/6-311+G(d) Opt Freq nosymm scf=xqc Geom=Check Guess=Read

O2 Singlet

0 1

Thank you very much Dr. Barsukov

Your response will be of immense help for me.