Location: Plant Polymer ResearchTitle: Chemometric Analysis of Multicomponent Biodegradable Plastics by Fourier Transform Infrared Spectrometry: The R-Matrix Method Author
Submitted to: Book Chapter
Publication Type: Book / Chapter
Publication Acceptance Date: 8/24/2011
Publication Date: 4/4/2012
Citation: Gordon, S.H. 2012. Chemometric Analysis of Multicomponent Biodegradable Plastics by Fourier Transform Infrared Spectrometry: The R-Matrix Method. Book Chapter. In Starch-Based Polymeric Materials and Nanocomposites: Chemistry, Processing, and Applications, Eds: Jasim Ahmed, Brijesh K. Tiwari and Syed H. Imam, CRC Press - Technology & Engineering, Chapter 13.
Technical Abstract: A new chemometric method based on absorbance ratios from Fourier transform infrared spectra was devised to analyze multicomponent biodegradable plastics. The method uses the BeerLambert law to directly compute individual component concentrations and weight losses before and after biodegradation of composite plastics. The method employs a new concept of absorbance ratios, the ratio matrix or R-matrix, that not only renders a multivariate system of Beer-Lambert law equations amenable to solution, but also affords the unknown absorbance coefficients from as few as two sample mixtures without external calibration. Unlike the classical K-matrix and the P-matrix chemometric methods, this R-matrix method does not require prior calibration against known composites and it does not require sampling ofmore different plastics than the number ofcomponents in the plastic being analyzed. Only two measurements, one before and one after biodegradation, are needed, particularly where spectra ofthe pure components are available. In this chapter the theoretical basis and derivation ofthe mathematical model for multi component systems and its solution algorithm are described. First, the validity of the R-matrix method is proved mathematically and by computer simulation ofmUlticomponent systems as defmed in the new R-matrix model. Then the method is validated experimentally by comparing test results with known biodegradable composites prepared in the laboratory, without dependence on or knowledge ofthe model. Use ofthe R-matrix model to analyze data containing inevitable experimental error revealed a potential for automatic infrared wavelength selection in software. The potential ofthe new R-matrix model is discussed as a chemometric method for analyses ofsolid-state biodegradable composites in general.