Jiang, Jian-Hui; Ozaki, Yukihiro; Kleimann, Michael; Siesler, Heinz W.:

Resolution of two-way data from on-line Fourier-transform Raman spectroscopic monitoring of the anionic dispersion polymerization of styrene and 1,3-butadiene by parallel vector analysis (PVA) and window factor analysis (WFA).

In: Chemometrics and Intelligent Laboratory Systems (Chemometrics Intellig.Lab.Syst.), Jg. 70 (2004) ; Nr. 1, S. 83-92
ISSN: 0169-7439
Zeitschriftenaufsatz / Fach: Chemie
Online Fourier-transform Raman spectroscopic monitoring of the anionic dispersion polymn. of styrene and 1,3-butadiene has been implemented with a noninvasive optic fiber Raman probe. After suitable pretreatment of the Raman spectra obtained, the resulting two-way data are analyzed using two self-modeling curve resoln. (SMCR) techniques, parallel vector anal. (PVA) proposed in the preceding paper, as well as window factor anal. (WFA), coupled with a slightly modified principal component anal. (PCA) procedure. The idea of PVA is to construct a set of subspaces comprising only one common (spectral) component, and then find a vector that is in parallel with a series of vectors coming from different subspaces. This procedure offers a versatile avenue to approach the unique resoln. of spectral profiles. The modified PCA procedure is a useful approach to eliminate the interference from nonreacting species and ext. the spectral information concerning only the active reactions. The results reveal that there are three Raman spectrally active species in the system, i.e., styrene, 1,3-butadiene and poly(butadiene), and the product, poly(styrene), turns out to have no Raman signals in the investigated spectral region. The spectral and the concn. profiles of the three species are resolved uniquely using the SMCR methods. The resolved spectral profiles exhibit only small discrepancies compared to the spectra measured for pure species, and the estd. concn. profiles coincide with the results predicted by previous copolymn. theory. These results demonstrate that the proposed PVA method and the modified PCA procedure are competitive approaches for the resoln. of two-way data from multivariate spectroscopic monitoring of reactions.