Near Infrared Spectra of Pellets and Thin Films of High-Density, Low-Density and Linear Low-Density Polyethylenes and Prediction of Their Physical Properties by Multivariate Data Analysis

The aim of the present study is to investigate in detail the near infrared (NIR) spectra of the three types of polyethylene, linear low-density polyethylene (LLDPE), low-density polyethylene (LDPE) and high-density polyethylene (HDPE), and to develop calibration models that predict their physical properties such as density, crystallinity and melting point. The effects of spectral resolution on the classification and the prediction of density for the three types of PE have been investigated. Furthermore, the NIR spectral differences among LLDPE, LDPE and HDPE have been explored in more detail using 2 cm −1 resolution. Principal component analysis (PCA) has been performed to differentiate the 18 samples of PE. They are classified into three groups, LLDPE, LDPE and HDPE, by a score plot of the PCA Factor 1 versus 3 based on the NIR spectra pretreated by multiplicative scatter correction (MSC). The 2 cm −1 spectral resolution yields a slightly better result for the classification. Partial least squares (PLS) regression has been applied to the NIR spectra after MSC to propose calibration models that predict the density, crystallinity and melting point of HDPE, LDPE and LLDPE. The correlation coefficient for the density was calculated to be 0.9898, 0.9928, 0.9925 and 0.9872 for the spectra obtained at 2, 4, 8 and 16 cm −1 resolutions, respectively, and the root mean square error of cross validation ( RMSECV) was found to be 0.0021, 0.0018, 0.0018 and 0.0023 g cm −3 , respectively. It has been found that the correlation coefficient and RMSECV for the prediction of the density and crystallinity change little with the spectral resolution. However, for the prediction of melting point, the higher resolutions (2 and 4 cm −1 resolution) provide slightly better results than the lower resolutions. NIR transmission spectra of thin films of LLDPE, LDPE and HDPE have also been investigated, and calibration models for predicting their density have been developed for the film spectra.