Improved accuracy for Raman spectroscopic determination of polyethylene property by optimization of measurement temperature and elucidation of its origin by multiple perturbation two-dimensional correlation spectroscopy

A novel strategy is demonstrated to improve the accuracy for determination of polyethylene (PE) density using Raman spectroscopy by optimizing the temperature of sample measurement. Spectral features associated with the conformation change of the polymer induced by temperature may provide valuable information to quantify important polymer properties such as density. To evaluate possible existence of an optimal temperature providing improved quantitative accuracy, Raman spectra of PE pellets with different densities were collected at eight different temperatures from 30 to 100 °C at 10 °C intervals. Using the spectral datasets collected at each temperature, partial least squares (PLS) models were developed using the reference PE density values determined by a standard density gradient method at 23 °C. Interestingly, the most accurate determination of density was realized at 70 °C. Multiple perturbation two-dimensional (MP2D) correlation analysis and differential scanning calorimetry (DSC) were used to examine the origin of improved accuracy at 70 °C. From these analyses, the pre-melt behavior of the PE samples was identified below their melting temperatures. Structural variations induced at the pre-melt stages enhance Raman spectral selectivity among the samples, thereby providing more accurate determination of PE density. The MP2D correlation analysis revealed the unforeseen thermal behavior of PE samples and successfully explained the improved accuracy at 70 °C.