Dear Sir,

Raman frequencies can be derived by using the keyword Freq=Raman. I think Polar=Raman is somewhat incorrect. please try the calculation with Freq=Raman keyword.

by

Karthick

@T. Karthick

Thanks for your answer, but my question is somehow different. I know I can use Freq=Raman: in this way I get both IR intensities and Raman activities with the same basis set; that is I compute the components of the polarization tensor at the same level of the force constants. However, it is possible, as mentioned in the paper I quoted, to start from Force constants obtained with a less extended basis set and compute polarization and Raman activities with larger basis sets. This procedure greatly improves the accuracy of the Raman spectra, reducing, at the same time, the total computational burden. As reported in the Gaussian09 manual, the keyword for such a procedure is Polar=Raman. However, this procedure requires also an excitation wavelength. Thus, I need someone who has used this procedure and can provide information,  especially on the choice of the wavelength.

Thank you

Stefano

Hai, 

        I hope you got an answer, otherwise just use the following procedure. 

Keyword for running a Raman calculation for different wavelength in Gaussian 09 is

CPHF=RdFreq which will add in the Route section.  Then, give one line space after the molecular specification and specify the desired frequency in Hatress or use nm for frequency in wavelength. 

For example,

opt+freq CPHF=RdFreq

.

.

Molecular specification

580 nm

Thank U

Dear Sir,

Have you solved this problem? I am trying to do 2-step Raman calculation, with freq=raman 1st and then polar=raman 2nd, but Gaussian complains "FileIO on non-existent file" at the 2nd calculation.

Thank you.