Determination of melt flow index and polymer additives in polyethylene based on IR spectra and multivariate modeling

Abstract

Protection against degradation is essential for preserving the favorable physical and chemical properties of polymers. There are several types of stabilizers employed as antioxidants, but the latest studies showed that artificial stabilizers spark environmental and healthcare concern. Here, we have used multivariate modeling based on ATR and transmission infrared spectra to provide information about the degradation and the applied additive stabilizer for polyethylene samples. The classification of samples based on the used additives was successful in single and combined additive groups as well, with accuracies over 0.9 in each case. ATR-IR combined with principal component analysis (PCA) and correlation and regression tree (CART) methods yielded the best classification model. Melt flow index values (MFI), which can correlate with the degradation and stability of the polymers, were predicted with partial least squares (PLS) regression and artificial neural network (ANN) models. The best combination for regression modeling was the transmission IR spectra with the ANN algorithm. The ANN model could predict the MFI values remarkably well, with an R2 value of 0.95, even for the test validation. Infrared spectroscopy coupled with chemometrics is suitable to be integrated into the quality control of polymers.