To convert epsilon to absorbance, you need to know length and concentration (Beer-Lambert law). I think it is easier to convert the experimental transmission or absorbance to epsilon because the length and concentration is usually known.

dear sir, thanks for your valuable answer, but i have theoretical IR values(i.e. frequency vs epsilon or intensity) from guassian 03 computational software , i have seen many journals(for example -journal of spectrochimica) that they are plotting theoretical IR values in transmission or absorbance vs frequency and compare with experimental IR values.It is difficulty to me. pls let me know to how convert this values

To transform the IR intensity measured in epsilon you need to provide the normalization procedure, i.e. you need to present calculated epsilon values in percentages. To do it you need to found the epsilon max. in all region and to divide all epsilon values by epsilon max. Resulting you have "absorbance vs. frequency" spectrum in which the epsilon values recalculated into the relstive absorbance.

Mr. Kumar,

useful software for this purpose is GaussView.

It may be found in http://www.gaussian.com/g_prod/gv5.htm.

GaussView operates with *.out file and visualizes as well 1H, 13C-NMR, Raman, NBO data, MOs and more. It is really a good software for visualization (and not only for visualization) of your theoretical data in publs.

Please find as attchment an example how looks a theoretically predicted IR-spectrum by Gaussian.

Dear Bojidarka B. Ivanova;

Can you take the procedure how can I do this in GaussView;

Best Regards 

you can calculate IR spectrum with Gaussian from properties , and you can convert absorbance to transmission by click option convert abs. to tra.

Be aware of the fact, that the Beer-Lambert law actually only describes propagation in but not transmission through a medium. A correct calculation of the latter does require the full dielectric function (tensor).

How can we get the exact value of epsilon for the peaks in UV spectra?

Depends. What kind of spectra did you record (transmission or reflection spectra? If reflection, internal, external or diffuse? If transmission, layer or liquid sample? If layer, what kind of substrate, if liquid, what is the thickness of the cuvette etc.). In other words, to help you we would require much more information...

its an absorption spectra obtained using gaussian 16, using TD-DFT methods. from the plot I want to obtain the epsilon values of the peaks. is there a way to find it using gaussview. I tried finding in .log file. could not find anywhere.

Gaussian should be able to give out either the functions of the absorption index, or the absorption coefficients etc. It could not calculate an absorption (absorbance?) spectrum otherwise. I hope someone familiar with Gaussian will help you.

the UV-vis plot obtained is epsilon vs wavelength plot. I tried using the equation from the gaussian page of UV-Vis spectra, but the value deviates from the value from the plot. Thanks Thomas MayerHofer.

The epsilon as computed in IR spectra has units M-1 cm-1. This epsilon = A/cl = Absorbance /concentration*length (M means molarity). So, normalizing A or epsilon (both normalize in similar fashion) will help finding T =1-A. %T can be computed likewise