Measurement of the glass transition temperature of an epoxy resin using principal components of Raman spectra

Many fibre reinforced polymer (FRP) composite manufacturing processes benefit from the ability to monitor the polymerisation of the resin constituent, either for process control or quality assurance purposes. Whilst most available resin monitoring systems achieve this by measuring the percentage conversion of reactants, measurement of the corresponding glass transition temperature, Tg, is often more informative. The ability to measure Tg directly from spectra acquired by Raman spectroscopy, using principal components analysis, has been demonstrated here for an amine-cured epoxy resin system. A regression model was generated using principal components identified from spectra acquired of cured resin samples produced at a range of isothermal conditions, regressed against corresponding Tg values measured by dynamic scanning calorimetry (DSC). Spectra acquired during the cool-down period following completion of one trial were incorporated into the calibration dataset to ensure effects of sample temperature were accounted for in the principal components analysis (PCA). Measurements obtained with the resultant model, generated using 5 principal components accounting for 97% of the variance in the calibration dataset, were shown to agree well with final sample DSC measurements from in situ cure trials, differing by <= 0.6. degrees C, and exhibiting negligible sensitivity to sample temperature. The results demonstrate the potential of Raman spectroscopy coupled with PCA as a non-contact cure monitoring technique, able to directly measure Tg of a thermosetting epoxy resin system typical of those used in FRP composites manufacture, independent of sample temperature.