I am a new user of Multiwfn (3.8 dev), first I would like to thank you a lot for this amazing software!

At the moment I am interested in charge transfer phenomena. I would like to reproduce the example on amino-nitro-biphenyl molecule (P2) as described in section 4.18.3 of Multwfn manual in order to generate a nice electron density difference plot (as seen for instance in Theor Chem Acc 2016, 135, 173), but I am not sure about what I am doing as I could not reproduce your results. This is what I am doing (I use Gaussian 16).

1) Optimization of P2 in the ground state: #p opt freq PBE1PBE/6-31+g(d) scrf=(iefpcm,solvent=ethanol)
2) Single-point TDDFT on the optimized ground state geometry: #p td=(root=1,singlet) PBE1PBE/6-31+g(d) scrf=(iefpcm,solvent=ethanol)

then I convert chk files to fch, and then fch to wfn using multiwfn. After following the steps on p.798-799 of the manual, my results are very off.

So could you please tell me what I am doing wrong here? I don't have to optimize the geometry in the excited state as the two geometries must be identical, right? Additionnaly I have a few questions:

a) Why did you use the keyword "density" in the extP2.gjf file?
b) I tried the procedure with .fch files instead of .wfn files and did not see any difference. Are .wfn files needed in this instance?   

Thanks a lot in advance for taking the time to answer and have great day!

Cheers,

Meurglys