As a first guess try to use baseline-corrected peak intensity vs. aging time. If it seems meaningful, you can go further by using internal standard method (which is recommended anyway, to divide the peak intensity by the peak intensity of another peak which supposedly does not change during oxidative aging) and only the last step is curve fitting, deconvolution and peak area calculation.

Please calculate carbonyl and hydroxyl indexes. Area under the peak in FTIR may work but due to aging, chances of getting multiple products are on higher side and this makes assumptions with errors.

If feel any difficulty, please let me know.

As a first guess try to use baseline-corrected peak intensity vs. aging time. If it seems meaningful, you can go further by using internal standard method (which is recommended anyway, to divide the peak intensity by the peak intensity of another peak which supposedly does not change during oxidative aging) and only the last step is curve fitting, deconvolution and peak area calculation.

Whether intensity or area, it's just numbers until you know the mechanism at play.  Unless some other mechanism is at play, I suspect your polymer can absorb some amount of water, and that the polymer is losing that water as it is heat aged?  Water of course has intensity where would expect O-H stretch, and it has an H-O-H bending mode near the C=O stretch region, so your intensity would certainly be affected.  If your [C=O or O-H] to C-H ratio is decreasing with heat aging, then moisture loss from polymer is something to look into.

In principle, you can. Your spectra must be standardized.

FTIR is good for qualitative analysis but when it can also be used as quantitative tool, though bit controversial... 

Read this paper: DOI: 10.1021/ed069pA217

One more thing to consider in your scenario is 'polymer aging" is mostly a surface phenomenon.. 

The suggestions are already there. I can add that to measure the breadth of the peak area, you draw a vertical line from x-axis to peak. Go to  middle of the line and measure the breadth and you start comparing. This  is known as half peak width, which is used in damping characterization and other properties.

As mentioned above, FT-IR is very helpful for qualitative characterization of polymer. It must be very careful to do quantitative comparison.  For your experiments, the best way is to do in-situ FT-IR characterization.

@Sridhar Vadahanambi:  Good reference on FT-IR, agrees with my personal experience. Just a comment:  the reason that there is a lot of spread in the data (Fig 1 of the reference) is that FT-IR is actually very quantitatively accurate so that sample inhomogeneities are easily picked up.  So you have a big chance of seeing such variations for solids with multiple components.  On the other hand, you'll find that with solutions, that FT-IR is a remarkably quantitative technique.  So any scatter in data may often be due to the fact that the sample is not homogeneous, not that FT-IR is accurate (though that can also be the case if the signal-to-noise for the peak in question is weak).

You can use area instead of peak intensity. The questions are: what type of polymer do you age and under what conditions?

Thank you very much for all the comments.

The polymer is shellac. Please find attached file of the mechanism of aging. Based on the figure, the effect of aging (due to heating, or humidity) will result

1. O-H will decrease

2. C-O increase

3. C=O ester increase

4. C-H is relatively constant

I did the method as suggested by Prof Gyorgy.

May I know what is the units of area under the curve

Dr. Gyrogy has given the full description perfectly. I like to add that half peak width is used for calculation in DMA curve. You may try.

For Razali question, area under the peak does not have unit as Intensity is expressed in arbitrary unit

Dear Profs,

Thanks for your valuable information. May I know,  how to calculate the peak ratio of hydroxyl group and carbonyl group in a FT-IR spectrum?

Thanks & regards,

Dr Sudharsan