Identification of historical polymers using Near-Infrared Spectroscopy

Portability of instrumentation, non-destructive spectral acquisition and a wealth of information that can be extracted from the spectra are features that have made Near-Infrared Spectroscopy an attractive topic in the analysis of cultural heritage.In this paper, a classification rule was developed using 41 classes of 535 historical and modern polymers using Near-Infrared Spectroscopy. The usability and the quality of the prediction model were tested on a blind sample set of 32 historical polymers. 20 out of 23 samples (87%) which are in the database used for the development of the classification model were correctly classified by the model. Eight darkly coloured samples could not be classified due to their low reflectance. The results of Near-Infrared Spectroscopy are compared to Attenuated Total Reflectance Fourier-Transform Mid-Infrared Spectroscopy and Pyrolysis-Gas Chromatography/Mass Spectrometry. Advantages and limitations of these techniques for the identification of historical polymers are discussed.