To normalize FTIR spectra, you typically divide the intensity values of each data point by a selected reference peak's intensity. This helps in comparing the relative changes in peak heights across different spectra. Choose a peak that remains relatively constant across all temperatures to serve as the reference peak. Common reference peaks include C-H stretches or specific absorption bands that are known to be stable under your experimental conditions.

As for which peak to use for comparison, consider using a peak that is not affected by the transformation from poly(amide ester) to polyimide. This will ensure that any changes observed in other peaks are due to temperature variations rather than the transformation itself. Look for a peak that remains distinct and consistent across your temperature range. This could be a characteristic peak of the polymer backbone, a functional group, or an impurity peak.

Keep in mind that selecting the right reference peak is crucial, as it will impact the accuracy of your comparisons. Additionally, it's a good practice to report the specific wavenumber or wavelength of the selected peaks in your analysis.

please, send your initial spectra of both polyamid and polyimid ( pictures) for undestanding how it is better to do?