Multiwfn forum

Dear Tian Lu,
I am trying to analyze the hole and electron distribution via Multiwfn. I do an optimization of the excited state of choice (1). Then using the .chk from (1) I do either a re-optimization including the density and IOp keywords (2), or a frequency analysis again with Density+IOp (3). The route sections of the calculations follow:

(1) #p opt td=(read,root=8,nstates=40) cam-b3lyp/cc-pvdz scrf=(solvent=water,read,pcm) nosymm empiricaldispersion=gd3bj guess=read geom=check
(2) #p opt td=(read,root=8,nstates=40) IOp(9/40=4) density=current cam-b3lyp/cc-pvdz scrf=(solvent=water,read,pcm) nosymm empiricaldispersion=gd3bj guess=read geom=check
(3) #p freq=(raman,savenm) td=(read,root=8,nstates=40) IOp(9/40=4) density=current cam-b3lyp/cc-pvdz scrf=(solvent=water,read,pcm) nosymm empiricaldispersion=gd3bj guess=read geom=check

I load the .fchk and .log files of each of the 3 calculations to Multiwfn and do the following for a fast Hole-Electron distribution visualization:
18 > 1 > 8 (for the root) > 1 > 2 (medium grid) > 3
Each of the 3 calculations gives me a different Hole-Electron picture for root 8. Am I correct to assume that (1) shows me the hole/elec distribution of root=1 due to the lack of the density=current keyword? But I am also puzzled by the discrepancy between (2) and (3), shouldn't they display the same result?

Thanks for any help on clarifying this point. Software used are Gaussian 16 revC.01 and Multiwfn 3.8(dev), release date: 2022-Jun-6

Hi,
The examples in p.706-707 of the manual (N-phenylpyrrole, NH2_C8_NO2) crash on the Windows v3.8 when you try to "Symmetrize the resulting TDM". They work fine in v3.7.