Determination of the blending parameters of ternary blends containing a terpolymer as a component copolymer by multivariate analysis of NMR data

Heat-resistant ABS resin, an industrial polymer material, is generally prepared by mixing AN/MS bipolymer and/or AN/ST/MS terpolymer with the matrix of ABS resin, where AN, ST and MS denote acrylonitrile, styrene and α-methylstyrene, respectively. In order to predict the thermal properties of ABS resins, it is important to know which copolymers are blended and in what ratio. In this study, statistical structural analysis was performed on polymer blends consisting of AN/ST bipolymer, AN/MS bipolymer and AN/ST/MS terpolymer to predict blending parameters for heat-resistant ABS resin. Nuclear magnetic resonance (NMR) spectral data of the polymer blends were used as explanatory variables to determine the chemical composition and mole fraction of the component copolymers. Partial least-squares (PLS) regression and least absolute shrinkage and selection operator (LASSO) regression were used as the exploratory techniques. The use of 1H NMR spectral data showed poor prediction, probably because of low resolution due to the narrow spectral width. Accordingly, 13C NMR spectra, which have wider spectral widths, were used, resulting in the successful prediction of the blending parameters of the ternary blends containing a terpolymer as a component. It should be noted that it was not necessary to obtain the NMR spectra of all the polymer blends as the explanatory variables, i.e., linear combinations of the NMR spectra of the component copolymers were sufficient.

論文本文は2026-08-05以降公開予定

© 2024. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/