I don't know why. This observation should be independent of choice of molecular structure.

Hi Tian,thank you for your quick answer.
Maybe I found a way to get also the vibronic contribution.
Do you know if I can get the dynamic polarizability only for the ground state?
Because I slightly modify the molecule along one normal mode but now Gaussian gives me back only the alpha static without also computing beta, even if I put CPHF and polar keyword.
Do you know why?

Dear Francesco,

Yes, vibrational contribution is not considered in the SOS calculation in Multiwfn.

Best,

Tian

Hi Tian,

firstly let me thank you for this super helpful program.
I am trying to calculate SFG and SDG hyperpolarizability with one frequency in IR, but Gaussian allows only to address the SHG hyperpolarizability.
Can I use the function number 19 of the Multiwfn hyperpolarizabilities menu "scanning w1 and w2 of beta" to get it,scanning the IR frequency?How much is accurate for small molecules?
Because I run it and it gave me back a really simple curve for the IR dependency, I think it takes into account only the electronic contribution to beta, neglecting the vibronic resonant ones, am I right? Could you suggest me a way to include theme?
Best regards,
Francesco

I have fixed this problem and updated Multiwfn 3.8(dev) on Multiwfn website, please download the latest version and retry.

I needed to calculate the SOS with non equilibrium salvation.
I used the following input:

%nosave
%nprocshared=40
%chk=g22_sos_bc.chk
%mem=40GB
#p cam-b3lyp/6-311++G(d,p) SCRF=(iefpcm,Solvent=dmso,NonEquilibrium=Save)

Solvente dmso

0 1
16       1.540108000      3.744434000      2.898043000
8        3.837623000      6.411342000      1.126905000
8        8.890953000      7.199642000      2.094155000

--link1--
%rwf=/Storage01/TempG09/clodoaldo/g22_sos_bc
%int=/Storage01/TempG09/clodoaldo/g22_sos_bc
%d2e=/Storage01/TempG09/clodoaldo/g22_sos_bc
%nosave
%nprocshared=40
%chk=g22_sos_bc.chk
%mem=40GB
#p cam-b3lyp/6-311++G(d,p) td=(nstates=300,root=1) IOp(9/40=5) Geom=Check Guess=Read SCRF=(iefpcm,Solvent=dmso,ExternalIteration,NonEquilibrium=Read)

Solvente dmso

0 1

but it works with 1500 states in multiwfn and not 300

Please careful check prompts on screen when Multiwfn export the files, the meaning of each column is also shown explicitly.

very grateful for the answer finally i can compare my values with the experimental values, and i want to comment a doubt in the meaning of the columns of the file beta_n because at difference with the file beta_n_comp in this file the meaning of each column appers but in the file beta_n don't appers i tried open with txt and opera (adjoint picture)

The "Projection of beta on dipole moment" can be determined by EFISHG (electric field induced second harmonic generation) experiment, therefore if your have EFISHG experiment data at a frequency w, you should calculate SHG form of beta at w and then compare the "Projection of beta on dipole moment" with experiment.

Note that the static beta, namely beta(0;0,0), cannot be determined experimentally in principle.

Hi, first i want to thank for the program in the last days this program and the manual has helped me a lot, and i want to ask about, which value can i use of my output of SOS to compare with the experimental value? in the paper are reported beta(zero) in 9 esu (they extrapolated the experimental value measure at 1580 nm), i have in my output of SOS of beta(0,0) the next data (adjoint picture), but i don't know which value i have to converted to esu and copare with the experimental value

This is correct if current values are in a.u.

Using the SOS model, in the alpha output file (average linear polarizability) to convert the values to esu just multiply by 1.4819E-25 and in the case of gamma by 5.03670E-40.

Is this information correct?

Current latest version of Multiwfn is able to perform two-state model and three-state model analysis for beta, see Section 3.200.8.2 of manual for introduction and Section 4.200.8.2 for analysis example.

Dear sir
   Some people are using 3 state model, two state model . Is it possible to get those results using Multiwfn . Look forward to hear from you.

  Thank you

If you want to use Gaussian to directly calculate gamma, see "examples\polar\NH3_gamma.gjf" in Multiwfn package for example. See Section 4.200.7 of Multiwfn manual for example.

If you want to use Multiwfn to calculate gamma based on SOS formalism, do not specify "polar" keyword, in current case you should only specify "TD(nstates=x)" and "IOp(9/40=5)". See Section 4.200.8 of Multiwfn manual for example.