Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly

Re^3: Transformation Firefly IR intensities to material property values (e. g. like absorption coefficient in 1/cm or 1/(cm*mol))

Alexey A. Popov

Dear Thomas,

unless you have experimentally determined absolute values of absorption (which means that you know optical path, concentration, beam square, etc), the play with absolute values from computations makes little sense. A standard way to compare experimental and computed IR spectra is simply applying an arbitrary uniform peak width, usually something like 2-4 cm-1, but you may adjust this value simply by choosing something that fits your experimental data better. And since nobody's usually working with absolute intensities, visualization packages you mentioned usually give the data in relative values (say, by normalizing to the highest intensity peak).

If you are still interested in the absolute values, then you can get a relation between the area and the value at the peak maximum simply from the properties of the functions (Gaussian, Lorentzian, etc) you use to approximate your peaks.

Say, for a Gaussian function y = A*exp{-(x-b)^2/2c^2}, area is A*c*(2*pi)^0.5

All the best,

On Wed Jul 27 '11 0:13am, Thomas H. wrote
>Dear Alex,

>Thanks for your answer and your references. But I think, I lose sight of something.

>I get the (centre) frequency and a value for the integrated intensity (independence from converting into other units) from the simulation – Am I wrong? That means I get a value for the area under a real curve (area of an absorbance peak), among other point of views because of the equation: integrated intensity = constant * integral of the absorbance from the lower to the upper peak frequency (wave number) - right?

>So, just from the simulation’s point of view I only now the centre frequency and the integrated intensity (area of a absorption band). I do not have information about the width of the peak, which I need e.g. for calculating the max. peak intensity by using a Lorentz or Gaussian fit. Are there standards for a reasonable peak width, to get absolute (estimated/approx.) values for the absorbance/absorption coefficient?

>Thanks again,


[ Previous ] [ Next ] [ Index ]           Wed Jul 27 '11 11:42pm
[ Reply ] [ Edit ] [ Delete ]           This message read 938 times