Multivariate analysis of 13C NMR spectra of branched copolymers prepared by initiator-fragment incorporation radical copolymerization of ethylene glycol dimethacrylate and tert-butyl methacrylate

In this paper, we report chemometric approach for structural analysis of branched copolymers. To evaluate chemical compositions and degree of branching (DB) values in branched copolymers, multivariate analyses, such as principal component analysis (PCA) and partial least-squares (PLS) regression, were applied to the 13C nuclear magnetic resonance (NMR) spectra of the carbonyl carbons of the copolymers prepared by initiator-fragment incorporation radical copolymerization of ethylene glycol dimethacrylate (EGDMA) and tert-butyl methacrylate (TBMA) with dimethyl 2,2’-azobisisobutyrate (MAIB). PCA successfully extracted information on monomeric units, such as EGDMA units, TBMA units and MAIB fragments, the last of which were incorporated via initiation and primary radical termination. The chemical compositions and the DB values of the copolymers were predicted
by PLS regression. Proper selection of a training set was found to be important for the prediction: the training set has to contain branched copolymers along with poly(EGDMA) and poly(TBMA). PLS regression using the appropriate training set allowed us to predict quantitatively the chemical compositions and DB values, without any assignments of the individual peaks.